CONTENTS
Section A: ESSENTIALS
Section B: ELECTRONIC BACKGAMMON: VS OTHER HUMANS
Section C: ELECTRONIC BACKGAMMON: VS MACHINE
Section D: RESOURCES
Section A: ESSENTIALS
``Backgammon is an obstacle race between two armies of 15 men each,
moving around a track divided into 24 dagger-like divisions known as
points.''
..The Rules
``It's just a game.''
-- Many
``Sport of mind.''
.. Alberto da Pra, President of WBF - Worldwide
Backgammon Federation
``It's a game of skill and luck.
When I win I can claim it's due to my good skill.
When I lose I can claim it's due to my bad luck.''
-- submitted by David Forthoffer davidf@lpd.sj.nec.com
``Backgammon is one of the oldest games in existence, dating back some
5000 years and believed to have been developed by the ancient Egyptians.
It is not a game of luck as many believe, but a strategic game of war;
in many ways as difficult to master as chess or Go. A random element
(luck) is certainly involved, but a champion player also uses the laws
of probability, intuition, imagination and psychology to outwit his
opponent''.
-- From the foward of the Expert Backgammon (Mac)
documentation.
``There's an aesthetic to the game, a flow. People think the game consists
primarily of math --- calculating odds and so forth. That's not true.
It's essentially a game of patterns, a visual game, like chess. Certain
patterns fit together harmoniously, make sense in a away that is
nontrivial.''
-- Paul Magriel
Answering ``Why do you play backgammon'':
``We have become a spectator society, one that experiences excellence
and creativity only by watching it on television or by reading about
it in newspapers or magazines...Perhaps the best way of becoming
something more than a spectator is to pursue activities that do not
receive mass media coverage. We can invent our own art forms, or at
least re-label existing forms as art. Backgammon, though it is very
old and very common, is an excellent art form. Patterns of points and
blots undergo poignant mutations. The player strains to work with them,
to control them. One's identity is not entirely intrinsic, nor is it
purely acquired. We can shape ourselves just as we can shape our
surroundings. By playing backgammon, that is - by creating patterns of
blots and points - I help to shape my identity, I set myself apart from
the spectators. I become alive.''
-- Felix Yen (from Anchors, Jan 92)
- A Backgammon board or layout.
- Thirty round stones, or checkers, 15 each of two different colors,
generally referred to as `men'.
- A pair of regular dice, numbered from 1 to 6. (For convenience,
two pairs of dice, one for each player, are generally used.)
- A dice cup, used to shake and cast the dice. (Again, it is more
convenient to have two dice cups.)
- A doubling cube---A six-faced die, marked with the
numerals 2,4,8,16,32 & 64. This is used to keep track of the number of
units at stake in each game, as well as to mark the player who last doubled.
Backgammon is an obstacle race between two armies of 15 men each,
moving around a track divided into 24 dagger-like divisions known as
``points''.
The Backgammon layout is divided down the center by a partition,
known as the ``bar'' (See Diagram 1), into an outer and inner (or home)
board or table. The side nearest you is your outer and home tables;
the side farther away is your opponents outer and home boards. The
arrows indicate the direction of play.
For purposes of convenience we have numbered the points in the
diagram. Though the points are not numbered on the actual board,
they are frequently referred to during play to describe a move or a
position. Your (X's) 4-point or 8-point will always be on your side
of the board; your opponent's (O's) will always be on his side of
the board.
A move from your 9-point to your 5-point is four spaces (the bar
does not count as a space). A move from White's 12-point to your
12-point, though it crosses from his board to yours, is but one
space, for these two points are really next to each other.
Diagram 2 shows the board set up ready for play. Each side has
five men on his 6-point, three men on his 8-point, five men on his
opponent's 12-point, and two men, known as ``runners'', on his opponents'
1-point. The runners will have to travel the full length of the
track, the other men have shorter distances to go. Note that play
proceeds in opposite directions, so that the men can be set up in two
ways. Turn the diagram upside down to see the layout if play were
proceeding in the other direction.
+-------------------------------------------------->
|
| +-----------------------------< X moves this direction
| |
| |
| | 13 14 15 16 17 18 19 20 21 22 23 24
| | +------------------------------------------+
| | | . . . . . . | | . . . . . . |
| | | . . . . . . | | . . . . . . |
| | | . . . . . . | | . . . . . . |
| | | . . . . . . | | . . . . . . |
| | | . . . . . . | | . . . . . . |
| | | | | | +----+
^ v | Outer Board |BAR| Home Board | | 64 |
| | | | | | +----+
| | | P O I N T S | | . . . . . . | Doubling
| | | . . . . . . | | . . . . . . | Cube
| | | . . . . . . | | . . . . . . |
| | | . . . . . . | | . . . . . . |
| | | . . . . . . | | . . . . . . |
| | +------------------------------------------+
| | 12 11 10 9 8 7 6 5 4 3 2 1
| |
| +---------------------------------------------->
|
+---------------------------------< Y moves this direction
Diagram 1 (Numbered from X's point of view)
13 14 15 16 17 18 19 20 21 22 23 24
+------------------------------------------+
| X . . . O . | | O . . . . X |
| X O | | O X |
| X O | | O |
| X | | O |
| X | | O | +----+
| |BAR| | | 64 |
| O | | X | +----+
| O | | X |
| O X | | X |
| O X | | X O |
| O . . . X . | | X . . . . O |
+------------------------------------------+
12 11 10 9 8 7 6 5 4 3 2 1
Diagram #2 (Numbered from X's point of view)
The object of Backgammon is for each player to bring all his men
into his home board, and then to bear them off the board. The first
player to get all his men off the board is the winner.
Each player casts one die. The player with the higher number makes
the first move, using the two numbers cast by his die and his
opponent's. In the event that both players roll the same number, it
is a standoff and each rolls another die to determine the first move.
In the event of subsequent ties, this process is repeated until the
dice turn up different numbers. (In some games, players double the
unit stake automatically every time they cast the same number; others
limit the automatic doubles to one. In tournament play, there is no
such thing as an automatic double.)
Each player's turn consists of the roll of two dice. He then moves
one or more men in accordance with the numbers cast. Assume he rolls
4-2. He may move one man six spaces, or one man four spaces and
another man two spaces. Bear in mind that, when moving a single man
for the total shown by the two dice, you are actually making two
moves with the one man---each move according to the number shown on
one of the dice.
If the same number appears on both dice, for example, 2-2 or 3-3
(known as doublets), the caster is entitled to four moves instead of
two. Thus, if he rolls 3-3, he can move up to four men, but each move
must consist of three spaces.
The players throw and play alternately throughout the game, except
in the case where a player cannot make a legal move and therefore
forfeits his turn.
A player makes a point by positioning two or more of his men on
it. He then ``owns'' that point, and his opponent can neither come to
rest on that point nor touch down on it when taking the combined
total of his dice with one man.
A player who has made six consecutive points has completed a
prime. An opposing man trapped behind a prime cannot move past, for
it cannot be moved more than six spaces at a time---the largest
number on a die.
A single man on a point is called a blot. If you move a man onto
an opponent's blot, or touch down on it in the process of moving the
combined total of your cast, the blot is hit, removed from the board
and placed on the bar.
A man that has been hit must re-enter in the opposing home table.
A player may not make any move until such time as he has brought the
man on the bar back into play. Re-entry is made on a point
equivalent to the number of one of the dice cast, providing that
point is not owned by the opponent.
A Player who has made all six points in his home board is said to
have a closed board. If the opponent has any men on the bar, he will
not be able to re-enter it since there is no vacant point in his
adversary;s home board. Therefore, he forfeits his rolls, and
continues to do so until such time as the player has to open up a
point in his home board, thus providing a point of rentry. It should
be noted, the he doesn't loses his turn, as he still retains the
ability to double his opponent before any of his opponents rolls,
assuming the cube is centered or on his side.
A player is compelled to take his complete move if there is any
way for him to do so. If he can take either of the numbers but not
both, he must take the higher number if possible, the lower if not.
[Another way of saying this...]
- If both parts of the roll can be played legally, then this must be
done. Note that you may play the roll in such a way as to move fewer
pips than the larger die indicates by playing the smaller die first ---
this is common in bearoff situations, and legal as long as each part of
the roll is played legally at the moment you play it.
- If only one part of the roll can be played legally,
then you must play the higher die if possible; if not, play the lower die.
--kw
Once a player has brought all his men into his home board, he can
commence bearing off. Men borne off the board are not re-entered into
play. The player who bears off all his men first is the winner. A
player may not bear off men while he has a man on the bar, or outside
his home board. Thus if, in the process of bearing off, a player
leaves a blot and it is hit by his opponent, he must first re-enter
the man in his opponents home board, and bring it round the board
into his own home board before he can continue the bearing off
process.
In bearing off, you remove men from the points corresponding to
the numbers on the dice cast. However, you are not compelled to
remove a man. You may, if you can, move a man inside your home board
a number of spaces equivalent to the number of a die.
If you roll a number higher than the highest point on which you
have a man, you may apply that number to your highest occupied
point. Thus, if you roll 6-3 and your 6-point has already been
cleared but you have men on your 5-point, you may use your 6 to
remove a man from your 5-point.
In some cases it may be advantagous to play the smaller die first
before applying the higher die to your highest point (See Compulsory
Move). For example, suppose you have one checker on your 5 point,
and two checkers on your 2 point. Your opponent has a checker on
the ace (one point) and on the bar. You roll 6-3. You may play the
3 to the 2 point then the 6 to bear a checker off the 2 point
leaving your opponent no shots (no blots for the opponent to hit).
The alternative, using the 6-3 to bear checkers off both the 5 and 2
points, would leave your opponent 20 out of 36 ways to hit your
remaining blot.
If you bear off all 15 of your men before your opponent has borne
off a single man, you win a gammon, or double game.
If you bear off all 15 of your men before your opponent has borne
off a single man, and he still has one or more men in your home board
or on the bar, you win a backgammon, or a triple game.
It is customary to cast your dice in your right-hand board. Both
dice must come to rest completely flat in that board. If one die
crosses the bar into the other table, or jumps off the board, or does
not come to rest flat, or ends up resting on one of the men, the dice
are ``cocked'' and the whole throw, using both dice, must be retaken.
The introduction of the doubling cube into the game is largely
responsible for the leap in popularity of modern backgammon.
Each face of the doubling cube bears a number to record
progressive doubles and redoubles, starting with 2 and going on to 4,
8, 16, 32 & 64. At the commencement of play, the doubling cube rests
on the bar, between the two players, or at the side of the board. At
any point during the game, a player who thinks he is sufficiently
ahead may, when it is his turn to play and before he casts his dice,
propose to double the stake by turning the cube to 2. His opponent
may decline to accept the double, in which case he forfeits the game
and loses 1 unit, or accept the double, in which case the game
continues with the stake at 2 units. The player who accepts the
double now ``owns'' the cube---which means that he has the option t
redouble at any point during the rest of the game, but his opponent
(the original doubler) may not. If, at a later stage he exercises
this option, his opponent is now faced with a similar choice. He may
either decline the redouble and so lose 2 units, or accept and play
for 4, and he now ``owns'' the cube. A player may double when he is on
the bar even if his opponent has a closed board and he cannot enter.
Though he does not roll the dice, for he cannot make a move, he still
has the right to double. Note that gammon doubles or backgammon
triples the stake of the cube.
From the FIBS help screens:
If you are playing an n-point match and your opponent is ahead
of you and he gets to n-1 points you are not allowed to use
the doubling cube in the next game to come
EXAMPLE:
5 point match
score
game # You opponent
1 0 3
2 0 4
3 1 4 (you were not allowed to double in this game)
4 3 4 (you were allowed to double again)
... ... ...
The Crawford rule is universally used in backgammon match play.
The Jacoby rule is used in money games. It states, that a gammon
or backgammon may not be scored as such unless the cube has been
passed and accepted. The purpose is to speed up play by eliminating
long undoubled games.
The Jacoby rule is never used in match play.
This rule applies to match games and states that in post-Crawford games
the trailer can only double after both sides have played two rolls. It
makes the free drop more valuable to the leader but generally just
confuses the issue.
Unlike the Crawford rule, the Holland rule has not proved popular,
and is rarely used today.
In money play, if player A doubles, and player B believes that he is
a favorite holding the cube, he may turn the cube an extra notch as he
takes, and keep the cube on his own side. For example, if A makes an
initial double to 2, B may, instead of taking the double and holding a
2 cube, say ``beaver'', turn the cube an extra notch to 4, and continue
the game holding a 4 cube.
If A believes that B's beaver was in error, some play that he may then
``raccoon'', turning the cube yet another notch (to 8 in the example). Cube
ownership remains with B. B may then if he wishes turn the cube yet
another notch, saying ``aardvark'', or ``otter'' or whatever silly animal
name he prefers (the correct animal is a matter of controversy), and so
forth.
Beavers and the rest of the animals may be played or not in money
play, as the players wish.
Beavers and other animals are never used in match play.
-- Andy Latto
It should be noted that the original cube turner can drop a beaver.
For example, suppose I miscount a bearoff and double, you accept and
say you want to beaver. I realize something is wrong and recount.
If I am horribly behind, I can drop the beaver, paying you the value
on the cube before you beavered.
-michael j zehr
A Chouette is a social backgammon variant for more than 2 players.
One player is ``the box'', and plays against all other players
on a single board. One other player is the captain, and rolls the
dice and makes the plays for the team that opposes the box. If the
box wins, the captain goes to the back of the line, and the next player
becomes captain. If the captain wins, the box goes to the back of the
line, and the captain becomes the new box.
Customs vary as to the rights of the captain's partners: In some Chouettes,
they may consult freely as to the way rolls should be played. In
others, consultation is prohibited. A compromise, where consultation
is allowed only after the cube has been turned, is popular.
Originally, Chouettes were played with a single cube. The only
decisions that players other than the captain were allowed to make
independently concerned takes: If the box doubled, each player on the
team could take or drop independently. Today, multiple-cube Chouettes
are more popular; each player on the team has his own cube, and all
doubling, dropping, and taking decisions are made independently by all
players.
-- Andy Latto
Single checkers (blots) on a point are vulnerable to enemy attack and must
start over if hit by n opponent's checker. Two or more checkers on a point
are safe from attack and can also be used for blocking or trapping your
opponent.
Essentially backgammon is a race to see who takes off all of his checkers
first. However, the shortest distance between two points is not always a
straight line. Most beginners, rarely leave exposed checkers and hit as
often as they can. As you will find out, this seemingly logical approach
is not the best strategy. The following is a simplification of some of
the factors that you should consider in forming a winning game plan:
- Distribution.
- Distribution is how evenly your checkers are divided among the points
occupied. It is usually better to have 3 checkers each on two different
points rather than 4 checkers one and 2 on the other. You should rarely
have six checkers on a point and almost never have any more. A player
with even distribution will seemingly get "luckier" dice than his less
flexible opponent.
- Exposure.
- Don't be afraid to leave shots early in the game to establish a strong
offense or defense. Be more cautious as your enemy's home board gets
stronger. The more points he has in his home board, the more difficult
it will be for you to re-enter after being hit. Conversely, the more
points that you control in your enemy's home board (anchors) the bolder
you may play. Even if his board is weak, limit the number of blots
(single checkers) to no more than four. If you are significantly ahead
in the race or position, then restrict your exposure to maintain your
lead.
- Blocking and Priming.
- Try to build points without gaps between them directly in front of the
enemy checkers in your home board to prevent their escape. Establishing
these critical points as early as possible in approximate order of
importance: 5, 4, 7 to start your blockade. Six points in a row is
called a prime. This makes it impossible for your opponent to escape
for as long as you can maintain that structure.
- Hitting.
- Try to hit checkers that are the most advanced or checkers that your
opponent would like to cover to establish an important point. Attack
only when it is advantageous to do so. For example, if you already
have two enemy checkers on the bar, it is more critical to make another
point in your home board than to hit a third checker. Also refrain from
hitting if it makes you more vulnerable than your opponent. Keep your
objectives in mind and don't be side-tracked. However, there is an old
backgammon adage that still carries weight,
"When in doubt, hit."
- Anchoring.
- Anchoring is establishing a defensive point (anchor)
in your enemies home board. This gives you a landing spot to come in on should
you get hit and prevents your opponent from making his home board. Early in
the game try to establish anchors on the higher points (20,21). If you become
significantly behind in the race, the lower points (22,23,24) have more value
as your strategy is to build your home board and wait for a shot. If you have
two anchors try to keep them on adjacent points.
These are just a few ideas for the beginner to get started and is not meant
as a tutorial. There are many fine books available if you awant more
information.
From Macintosh Expert Backgammon Documentation
by Tom Johnson komodo@netcom.com
From: kwoolsey@netcom.com (Kit Woolsey)
Subject: Re: What are the best ways to play the opening rolls?
Date: Sun, 19 Mar 1995 05:19:43 GMT
[...]
Now, on the what I believe is an accurate synopsis of the 15 possible
opening rolls:
2-1: The slotting play 13/11, 6/5 and the splitting play 24/23, 13/11,
the two most common plays, seem to be about equal. Nothing else is a
serious contender.
3-1: 8/5, 6/5 is obviously the only play.
4-1: The splitting play 24/23, 13/9 has come out clearly superior to the
slotting play 13/9, 6/5. Probably the reason is that with the builder on
the 9 point there are so many good pointing numbers next turn anyway that
you don't need the 5 point slotted.
5-1: The splitting play 24/23, 13/8 has come out a bit better than the
slotting play 13/8, 6/5. A third less common alternative, 24/18, came
out clearly worse.
6-1: The obvious 13/7, 8/7 is correct. Magriel's experiment of 13/7, 6/5
is awful.
3-2: The splitting play 24/21, 13/11 came out a bit better than building
with 13/10, 13/11.
4-2: 8/4, 6/4 of course.
5-2: The normal play for years has been 13/11, 13/8. However the newer
splitting play, 24/22, 13/8, (shunned because of the crushing 5-5 threat)
has come out a bit better. The slotting play of 13/8, 6/4 (which used to
be my choice) did not survive the rollouts -- it was clearly inferior.
6-2: The splitting play of 24/18, 13/11 comes out fairly clearly
superior. Running with 24/16 is 2nd, but the run isn't far enough.
Slotting with 13/5 (a common choice several years ago) was definitely in
third place.
4-3: The building play of 13/10, 13/9 and the common splitting play of
24/20, 13/10 were just about tied. The alternative split of 24/21, 13/9
was only a little behind.
5-3: The simple 8/3, 6/3 is clearly best. The once common 13/10, 13/8
has been found vastly inferior.
6-3: The splitting 24/18, 13/10 comes out best, but the running play of
24/15 is not too far behind.
5-4: Splitting with 24/20, 13/8 and building with 13/9, 13/8 come out
quite close (that builder on the 9 point is powerful), with the split
generally a tiny bit better. 24/15 is weaker still.
6-4: Both running with 24/14 and splitting with 24/18, 13/9 are about
equal. However the once laughed at 8/2, 6/2 has reared its head as a
serious contender and comes out about equal with the other choices --
nice play to try if you get familiar with it, since your opponent
probably won't be.
6-5: The simple 24/13 is clearly better than any other possibilities.
Section B: ELECTRONIC BACKGAMMON: VS OTHER HUMANS
On July 19, 1992 at 10:56:22 GMT, Marvin announced the birth of the FIBS.
FIBS is an abbreviation for First Internet Backgammon Server. It
is a server program written by Andreas Schneider
marvin@fraggel65.mdstud.chalmers.se. Praise to him!
After using a site in Aachen Germany, It moved to the present site ins Sweden
on December 3, 1993.
You can connect to FIBS using telnet; the server runs on machine
fraggel65.mdstud.chalmers.se (raw address 129.16.235.165). Be sure
to telnet to port 4321, because telnetting to the default port will
give you a regular unix login prompt, which will be of no use to
you since you probably have no account on that machine. For
instance, on a unix machine which is connected directly to the
internet, you issue one of the following commands:
telnet fraggel65.mdstud.chalmers.se 4321
telnet 129.16.235.165 4321
and then you are connected to FIBS.
Now you have to log in to FIBS. If you already have an account on
FIBS, you use the login name and password you selected. If you're
a first time user, you must log in as guest. Then you are granted
a limited kind of access, until you make yourself known by choosing
a FIBS user name and a password. From that moment on, you can use
FIBS to play against other players, human or otherwise, from all
over the world.
Here is an example connect and login sequence for new users:
%telnet fraggel65.mdstud.chalmers.se 4321
Trying 129.16.235.165 ...
Connected to fraggel65.mdstud.chalmers.se.
Escape character is '^]'.
WELCOME TO THE
_______ _ ______ _____
| _____| | | | __ \ / ____|
| |___ | | | |__| | | |____
| ___| | | | __ < \____ \
| | | | | |__| | ____| |
|_|irst |_|nternet |______/ackgammon |_____/erver
If something unexpected happens please send mail to:
marvin@fraggel65.mdstud.chalmers.se (Andreas Schneider)
Bug reports are welcome.
This server is on the net to meet people from all countries.
All sorts of racists and fascists are not allowed to login here!
Rude language will not be tolerated on this server. Be nice.
LOGIN AS guest IF YOU ARE NEW TO THIS SERVER!
One account per person only!
Friday, September 30 10:23:10 MET ( Fri Sep 30 09:23:10 1994 UTC )
login: guest
Welcome to FIBS. You just logged in as guest.
Please register before using this server:
Type 'name username' where username is the name you want to use.
The username may not contain blanks ' ' or colons ':'.
The system will then ask you for your password twice.
Please make sure that you don't forget your password. All
passwords are encrypted before they are saved. If you forget
your password there is no way to find out what it was.
Please type 'bye' if you don't want to register now.
ONE USERNAME PER PERSON ONLY!!!
> name Newbie
Please give your password:
Please retype your password:
You are registered.
Type 'help beginner' to get started.
>
Once logged in, you are wise to read the help screens of FIBS.
Read about how not to hear other people's shoutings, how the rating
system works, how to watch other people play, how to talk to other
people, how to invite people to play, and of couse, how to play.
Everything you need is in the help screens. One thing: if you
wish to read the help screens without logging into FIBS,
they have been made available to WWW by Mike Quinn at
http://www.cybercom.net/~damish/backgammon/mike_quinn/fibs.htm.
Mark Damish made a version available for ftp or online reading from
http://www.cybercom.net/~damish/backgammon/fibshelp.html
When you are a little accustomed to FIBS, you can enter
tournaments, which are organised occasionally by volunteers. Read
newsgroup rec.games.backgammon,
check out FIBS' login message, or listen for rumours spreading.
Also, if you like FIBS, it will pay you to take the trouble to
install/use a more friendly interface than a simple telnet client
program. Several of these are available, see section
Are there any GUI's for FIBS?.
FIBS description last updated on October 4th, 1994
by Vincent
Zweije (zweije@wi.leidenuniv.nl)
FIBS Command Help Summary. (One liners)
- about - display information about the server
- accept - accepting doubles and resigns
- address - make your email address known to other users.
- autologin - how the tinymud style autologin feature works
- average - show average number of users
- away - leaving a message for other users before leaving the
terminal
- back - back again after the away command was used
- beaver - offering an instant redouble that is a beaver
- beginner - very short introduction to the server
- blind - Stop people from watching you.
- board - displays the board again
- boardstyle - the various boardstyles
- bye - leave the first internet backgammon server. Aliases for
bye include: adios, ciao, tschoe, end, exti, logout, and quit.
- client - one way to use a client
- cls - clear the screen on a vt100 terminal
- commands - how commands are entered.
- complaints - how to complain about cheaters
- countries - where do the players live
- crawford - The Crawford rule
- date - equivalent to the time command
- dicetest - show statistics about the dice
- double - Ship that cube!
- erase - How and why accounts are erased
- formula - The formulas used to calculate rating changes
- gag - Inhibit yourself from hearing a players shouts etc...
- help - help on different topic
- hostnames - how to interpret hostnames given by the who command
- invite - invite another user to play a game of backgammon
- join - accept an invitation from another player
- kibitz - talking to players and watchers
- last - Display information about login times
- leave - leave and save a game
- look - Take a short look at a game
- man - alias for help
- message - Leave a message for a user
- motd - Display the message of the day
- move - Moving pieces on the board
- names - name completion
- off - bear off pieces with every possible move
- oldboard - Display the board of a saved game.
- oldmoves - Display the moves of a saved game.
- otter - Offering an instant redouble that is an otter
- panic - save a game to a special file
- password - change password
- pip - Display pip count
- raccoon - Offering an instant redouble that is a raccoon
- ratings - Display information from the rating list
- rawboard - how to interpret the raw board output
- rawwho - A version of the who command for client programs.
- redouble - accepting doubles by redoubling
- reject - Drop a double. Reject a resignation.
- resign - resign a game
- roll - roll the dice
- rules - The basic rules of backgammon
- rule1 - how the board looks like
- rule2 - the direction you move pieces
- rule3 - the goal of the game
- rule4 - rolling the dice
- rule5 - moving pieces
- rule6 - moving pieces
- rule7 - bearing off pieces
- rule8 - winning
- rule9 - doubling
- save - save your current toggle settings
- say - talk to your opponent
- screen - how to tell FIBS about your screen
- set - how to set variables that are not toggles
- shout - say something to all users
- show - Display information
- shutdown - shutdown the server (privileged users)
- sortwho - how the 'who' command sorts it's output
- stat - display system usage information about the server
- tell - say something to a specific player
- time - display the current time
- timezones - How the server supports different timezones
- tinyfugue - a few hints on using the TinyFugue client
- toggle - display or change the value of toggles
- toggle-allowpip - Enable/Disable the servers `pip' command.
- toggle-autoboard - Enable/Disable automatic board redraws.
- toggle-autodouble - Enable/Disable Auomatic doubles on the 1st roll.
- toggle-automove - Enable/Disable Automatic movement of forced rolls.
- toggle-bell - Enable/Disable the bell in talking or invites.
- toggle-crawford - Enable/Disable Crawford. Both players need to agree.
- toggle-double - Enable/Disable automatic rolling.
- toggle-greedy - Enable/Disable automatic bearoffs if possible.
- toggle-moreboards - Redraw every move, or every move and roll.
- toggle-moves - Enable/Disable listing of moves at end of game.
- toggle-notify - Enable/Disable server notification of players logging in and out.
- toggle-ratings - Enable/Disable the display of the rating calculation.
- toggle-rawboard - Replaced by set boardstyle <1..3>
- toggle-ready - Toggles wether you are ready to play games.
- toggle-report - Enable/Disable server messages when other players start or finish a match.
- toggle-silent - Enable/Disable hearing players shouts.
- toggle-telnet - Toggles extra newlines.
- toggle-wrap - Toggles whether you or the server
wraps lines larger than 80 charactors.
- unwatch - stop watching a player
- version - display version number of the server
- watch - watch a player
- wave - wave goodbye before leaving to players who receive shouts
- where - display full hostnames
- whisper - say something to watchers of a game
- who - display information about currently logged in users
- whois - Display information about a player
- !! - repeat the last command
[Last updated June 1995. Are there any missing commands?]
For more detailed information on FIBS commands, type ``help'' at the
while on FIBS or check out Michael Quinn's Guide to FIBS at:
http://www.abekrd.co.uk/FIBS
FIBS Rating Formula [From the FIBS man pages.]
- NAME
- formula - The formulas used to calculate rating changes
- DESCRIPTION
-
These are the formulas used to determine the ratings of a player:
Let's say that two players P1 and P2 were playing a n-point match.
The ratings of the players are r1 for P1 and r2 for P2 .
- Let D = abs(r1-r2) (rating difference)
- Let P_upset = 1/(10^(D*sqrt(n)/2000)+1) (probability that underdog wins)
- Let P=1-P_upset if the underdog wins and P= P_upsetif the favorite wins.< /LI><
/LI>< /LI>< /LI>< /LI>
- For the winner:
- Let K = max ( 1 , -experience/100+5 )
- The rating change is: 4*K*sqrt(n)*P
- For the loser:
- Let K = max ( 1 , -experience/100+5 )
- The rating change is: -4*K*sqrt(n)*P
The 'experience' of a player is the sum of the lengths of all matches
a player has finished. Every player starts with a rating of 1500 and
an experience of 0.
- SEE ALSO
- ratings
From:gmortens@newstand.syr.edu (Gerald E Mortensen)
Subject: fibs ratings formula plots
Date: 23 Dec 1994 22:28:16 GMT
i made these plots from the fibs ratings formula. experience >500. if
you can't read these try setting your font to fixed or courier.
P(win) vs. ratings difference
0.8 ++-----+-------+------+-------+------+------+-------+-----++
+ + + + + + + + C +
| : C |
0.7 ++ : C B ++
| : C B |
| : B |
| : C B |
0.6 ++ : C B A A ++
| : B A A |
| : B A A |
0.5 ++ A A A ++
| A A B : |
| A A B : |
0.4 ++ A A B C : 1 pt match A ++
| B C : 5 pt match B |
| B : 9 pt match C |
| B C : |
0.3 ++ B C : ++
| C : |
+ C + + + + + + + +
0.2 ++-----+-------+------+-------+------+------+-------+-----++
-400 -300 -200 -100 0 100 200 300 400
ratings change for a win vs. ratings difference
10 ++-----+-------+------+-------+------+------+-------+-----++
+ + + + + + + + +
9 ++ C : ++
| C C : |
8 ++ C : 1 pt match A ++
| : 5 pt match B |
| C : 9 pt match C |
7 ++ C : ++
| B C : |
6 ++ B B C ++
| B B : C |
5 ++ B : C ++
| B B C |
4 ++ : B C ++
| : B B C |
| : B B C |
3 ++ : B B ++
| A A A A A : |
2 ++ A A A A A A A A ++
+ + + + + + + A A +
1 ++-----+-------+------+-------+------+------+-------+-----++
-400 -300 -200 -100 0 100 200 300 400
ratings diff (your rating - opponent's)
ratings change is the same for both players if both
have experience > than 500 (or have equal experience < 500).
jay (wilfo)
FIBS - Rating Changes 2/16/95
change in rating when favorite wins
rate points in match
diff 1 2 3 5 7 9 11
------------------------------------------------------------
0 2.00 2.83 3.46 4.47 5.29 6.00 6.63
40 1.95 2.74 3.33 4.24 4.97 5.59 6.13
80 1.91 2.64 3.19 4.01 4.65 5.18 5.63
120 1.86 2.55 3.05 3.79 4.34 4.77 5.14
160 1.82 2.46 2.92 3.56 4.03 4.38 4.67
200 1.77 2.37 2.78 3.35 3.73 4.01 4.22
240 1.73 2.28 2.65 3.13 3.44 3.65 3.79
280 1.68 2.19 2.52 2.93 3.16 3.31 3.39
320 1.64 2.11 2.39 2.73 2.90 2.99 3.02
360 1.59 2.02 2.27 2.54 2.65 2.69 2.68
400 1.55 1.94 2.15 2.35 2.42 2.41 2.37
440 1.50 1.86 2.03 2.18 2.20 2.15 2.08
480 1.46 1.78 1.92 2.01 1.99 1.92 1.83
change in rating when underdog wins
rate points in match
diff 1 2 3 5 7 9 11
------------------------------------------------------------
0 2.00 2.83 3.46 4.47 5.29 6.00 6.63
40 2.05 2.92 3.60 4.70 5.61 6.41 7.14
80 2.09 3.01 3.74 4.93 5.93 6.82 7.64
120 2.14 3.10 3.88 5.16 6.25 7.23 8.13
160 2.18 3.19 4.01 5.38 6.56 7.62 8.60
200 2.23 3.28 4.15 5.60 6.86 7.99 9.05
240 2.27 3.37 4.28 5.81 7.14 8.35 9.48
280 2.32 3.46 4.41 6.02 7.42 8.69 9.88
320 2.36 3.55 4.53 6.22 7.68 9.01 10.2
360 2.41 3.63 4.66 6.41 7.93 9.31 10.6
400 2.45 3.72 4.78 6.59 8.17 9.59 10.9
440 2.50 3.80 4.89 6.76 8.39 9.85 11.2
480 2.54 3.88 5.01 6.93 8.59 10.1 11.4
ratio - points lost to points won by favorite
rate points in match
diff 1 2 3 5 7 9 11
------------------------------------------------------------
0 1.00 1.00 1.00 1.00 1.00 1.00 1.00
40 1.05 1.07 1.08 1.11 1.13 1.15 1.17
80 1.10 1.14 1.17 1.23 1.28 1.32 1.36
120 1.15 1.22 1.27 1.36 1.44 1.51 1.58
160 1.20 1.30 1.38 1.51 1.63 1.74 1.84
200 1.26 1.38 1.49 1.67 1.84 2.00 2.15
240 1.32 1.48 1.61 1.85 2.08 2.29 2.50
280 1.38 1.58 1.75 2.06 2.35 2.63 2.91
320 1.45 1.68 1.89 2.28 2.65 3.02 3.39
360 1.51 1.80 2.05 2.53 2.99 3.47 3.95
400 1.58 1.92 2.22 2.80 3.38 3.98 4.61
440 1.66 2.05 2.40 3.10 3.82 4.57 5.37
480 1.74 2.18 2.60 3.44 4.31 5.25 6.25
Batting Average for favorite to maintain rating
points in match
diff 1 2 3 5 7 9 11
--------------------------------------
0 .500 .500 .500 .500 .500 .500 .500
40 .512 .516 .520 .526 .530 .534 .538
80 .523 .533 .540 .551 .561 .569 .576
120 .534 .549 .560 .577 .590 .602 .613
160 .546 .565 .579 .602 .619 .635 .648
200 .557 .581 .598 .626 .648 .666 .682
240 .569 .596 .617 .650 .675 .696 .714
280 .580 .612 .636 .673 .701 .725 .744
320 .591 .627 .654 .695 .726 .751 .772
360 .602 .642 .672 .716 .750 .776 .798
400 .613 .657 .689 .737 .772 .799 .822
440 .624 .672 .706 .756 .793 .820 .843
480 .635 .686 .723 .775 .812 .840 .862
FIBS ratings tables submitted by William C. Bitting
btbr68a@prodigy.com
wbitting@crl.com
FIBS ratings reports are posted regularly to Rec.games.backgammon. Back issues are
available from:
http://www.columbia.edu/~radev/backgammon/fibsratings/
http://www.cs.ualberta.ca/~banks/fibs/test.html
Currently there several computer programs on FIBS:
tesauro (the original neural net bg program TD-Gammon)
mloner (neural net)
idiot (neural net (JellyFish))
jellyfish (nn)
loner (The 1-pt version of mloner)
EXBGthree
fatboy (nn)
fattest (nn)
jemina (Algrithmic, entering cocoon, to emerge as a nn)
music
Big_Brother (only logs matches)
Some programs play with humans entering the data, while others
are full fledged bots.
FIBS (and FIBS/W) Instruction book.
Hunter Jones has put together a very nice reference to FIBS and FIBS/W.
It is nicely typeset and printed on heavy paper stock. The contents
make a nice reference to FIBS and the FIBS/W interface. Commands are
pre-sorted by catagory, and it makes looking for an answer extremely
easy. It is 8 pages on 6 sheets of paper, and is especially worthwile
for the new player. It is not just a rehash of the man pages.
Price is $4 for U.S. addresses, $6 US for foreign addresses.
All payments must be in US funds (check, money order or cash).
If you wish expedited shipment, enclose suitable payment.
(For example, $10 additional for US FedEx overnight.) Be sure
to enclose your address (FedEx and the like cannot deliver to
PO Boxes.)
Contact Hunter Jones at: hunter@ix.netcom.com
6617 Struttmann Lane
Rockland MD
20852
From: thrash@mercury.interpath.net
Newsgroups: rec.games.backgammon
Subject: online FIBS help for OS/2 users
Date: 7 Mar 1995 03:45:57 GMT
For FIBS players who use OS/2:
I recently created an online help file (.INF) for all the FIBS
commands. It's basically the same help you get from the FIBS server
but with hyper-text links to related commands.
If you're new to FIBS, I'm sure it will help - although I can't
guarantee it will increase your FIBS rating ;)
Email me if your interested, I can send it via uuencode mail.
-J-
[This is copied verbatim, with permission, from OK.FAQ. References to 'OK'
are referring to the bridge server.]
[Permission from mclegg@cs.ucsd.edu (Matthew Clegg) for use here.]
In addition to having access to a Unix system, you must also be connected
to the Internet. The Internet is a worldwide computer network which was
founded for the sake of promoting research and education. Recently,
the Internet has been broadening its mission and it's likely that
soon the Internet will be open for commercial as well as educational
uses.
Already it is possible for the general public to obtain access to
the Internet for a modest fee in many metropolitan areas of the US.
A few representative Internet providers include:
Area Served Voice No. Email Organization
----------- -------- ----- ------------
West Coast 408-554-UNIX info@netcom.com Netcom Online Comm. Svcs
Boston 617-739-0202 office@world.std.com The World
New York City 212-877-4854 alexis@panix.com PANIX Public Access Unix
Many OKbridgers play from home using a PC or Mac and a modem.
Frequently, these people have obtained access to the Internet by
purchasing an account from a "public access Unix system connected to
the Internet," which is the jargon describing the service provided by
the above companies. Having obtained such an account, it is usually a
simple matter to obtain OKbridge and begin playing (see below).
If you will be searching for a means to use OKbridge, it is important
to remember the wording, "public access Unix system (directly) connected
to the Internet." There are a number of BBS operators who have Email
connections to the Internet, but this is not sufficient. Also,
there are several network services which provide access to the
Internet but which are not Unix based (Delphi is a notable example).
For more information about the Internet, which is an amazing and
wonderful resource, see the books:
Krol, Ed, The Whole Internet: User's Guide & Catalog,
O'Reilly & Associates, 1992.
Kehoe, Brendan P., Zen and the Art of the Internet: A Beginner's Guide,
2nd ed., Prentice Hall, 1993.
LaQuey, Tracy, with Jeanne C. Ryer, The Internet Companion:
A Beginner's Guide to Global Networking, Addison-Wesley, 1993.
These books are filled with useful information about Unix and the
Internet, including how to send electronic mail, how to download
free software, and how to access some of the many information services
which are available on the Internet.
The World Wide Web (WWW or 'Web' for short) is a system by which text,
pictures, audio files and movies can be transmitted across the internet.
Old resources you may have heard of -- telnet, news, gopher, ftp -- can
all now be regarded as part of the Web.
Many of the 'pages' on the Web are written in a language called HTML.
This language allows basic formatting of the text, and images to be
included within the text, but also it allows 'links' to other documents
which may be local or on the other side of the world. For example, I
could say 'I have information about cows' and the word 'cows' would be
highlighted somehow (underlined or in a different colour). If you select
that word -- typically by clicking your mouse on it or pressing Enter if
you have no mouse -- you will be taken to a page about cows which could
be another page of mine or of someone in Australia. It does not take much
imagination to see how I can then hop all over the world, following these
links and reading all manner of information.
In order to access the Web, you need a so called 'client program' or
'browser'. The two most popular are called Mosaic (for graphical terminals)
and 'lynx' for text based terminals. If you have got one of these programs,
you can start browsing the Web immediately. If not, do what you can to get
one!
[The Netscape browser is also now worth a mention. It was only in beta test
at the time I originally wrote this, and a bit buggy, but now it's my
preferred browser. You can get it via anonymous ftp from ftp.mcom.com in the
directory /netscape : it's free for academic and non-profit use.
S.T. 30/1/95]
Stephen R. E. Turner
e-mail: sret1@cam.ac.uk
WWW: http://www.statslab.cam.ac.uk/~sret1/home.html
Good places to find local internet providers are listed at the
following web sites:
http://www.internic.net
http://thelist.com
Tinyfugue is a telnet client program which breaks the screen into
separate 'panes' for input and output. A specialized version exists where
a non scrolling backgammon board is displayed in a third pane. The
specialized version is available for anonymous ftp from
figment.csee.usf.edu in the directory /pub/misc/FIBS_client.
The FIBS command "help tinyfugue" will provide some hints on using this program.
Patches made by:
David Eggert eggertd@aisb.ed.ac.uk
(window routines)
Andreas Schneider marvin@fraggel.mdstud65.chalmers.se
(board printing routines and /board command)
figment.csee.usf.edu /pub/misc/FIBS_client
From: jlehett@mailhost.gate.net
Newsgroups: rec.games.backgammon
Subject: OS/2 Native FIBS-Tinyfugue Client Available now!
Date: 11 Dec 1994 01:49:27 GMT
I just uploaded my port of the fibs-tinyfigue client to the incoming
directories of the hobbes and ftp-os2/cdrom OS/2 sites. It only works
via TCP/IP connections (SLIP/TIA are fine!) and requires the emxrt.zip
support, so be sure to get that too. The included text file tells about
the port so have a look all OS/2 fibsters!
Send along any bugs, etc, and I'll see what I can do.
----
John J. Lehett
Land-J Technologies
JLEHETT@GATE.NET
[ It has been reported (95-06) That this e-mail address is no longer in use]
xfibs - graphical interface to FIBS (First Internet Backgammon Server)
DESCRIPTION
XFibs is a Motif-based interface to FIBS. FIBS allows you to play backgammon
against other people (and an increasing number of computer programs). Unfortu-
nately, FIBS is text-based and moves must be typed in numeric notation. XFibs
graphical interface provides a mouse-driven board, which minimises text input.
Below is an explanation of the various feautures of XFibs, and what you may do to
further customize it according to your own desires. At the very end you'll also
find the backgammon rules.
HOW TO PLAY
XFibs draws two windows, one to display a backgammon board which is fully resizable
and scalable; and another for the text information. FIBS is a vibrant and lively
place, often with lots of banter going on as well as matches starting and finish-
ing. With XFibs you'll only really use the text window occasionally because you can
forget all about how the board is numbered, in XFibs you move your pieces with the
mouse.
The right button brings forward a popup-menu with
several choices like "roll dice", "double" etc., selecting "roll dice" when it's
your turn will cause two dice to appear.
The left button allows you to click on a piece and drag
it to where you want to put it. You can then release the left button to drop the
piece. If the move is valid, XFibs will draw the piece at the new location. (You
can now pick-up a piece and move both dice in one action: i.e. 24-13 with 6-5,
pick up from 24, drop on 13. The left-hand die is the default first die of a
move, the the right-hand die. If the left-hand die can't move, XFibs will try
the reverse combination. So If you have a roll in both combinations are legal,
but only one hits an opponent, you may need to drag-and-drop to guarantee a hit
or a miss).
Alternatively, you can double-click on the middle button
over a piece to have XFibs move it (again the left-hand die is the default first
die of a move).
If you decide that you didn't want to move a piece, you
can take it back (either by drag-and-drop or by popup).
Once you are happy with a move, it has to be sent to
FIBS. This is done by clicking your right mouse button again. This popup menu
changes according to what is going on. Now it says "accept move", "undo move"
etc. Between games in a match it says "join" "leave". Try it out!
If you get a text description of the board in the output
window, you have to issue a 'set boardstyle 3' to FIBS. You may then save your
setup my typing 'save' or include this command after a 'on_login' command in
your startup file. (see below)
All in all, just fool around with it, it isn't that hard
to figure out... (At least I hope it isn't)
[...]
AUTHOR
Torstein Hansen
Minor patches
made by David Eggert (Snoopy)
Changes since version 0.7
made by Mike Quinn (mikeq)
Available from: ftp://itekiris.kjemi.unit.no/pub
and/or ftp://www.abekrd.co.uk/pub/fibs/
The source is available from: ftp://ftp.x.org/contrib/games/xfibs08.tar.gz
MacFIBS greatly enhances the virtual backgammon experience; it's
backgammon played "The Macintosh Way".
MacFIBS provides a multi-window, graphical front end to
FIBS , vastly
superior to the "dumb terminal" telnet scrolling text format that FIBS uses
underneath. It also makes excellent use of sound to reinforce the backgammon
playing experience.
Rather than viewing backgammon positions as a series of
X's and O's in a crude character-based text window, MacFIBS offers a full color
backgammon board. Instead of typing cryptic commands like 'm 24 22 15 14', you
drag colored checkers around the board, exactly like playing a real game. The
user can select from two board sizes and choose which color and direction to
play. Real-time pip count information is also displayed.
Other windows include: a Player window to invite, get
info, or watch other players, an elegant Chat window for conversing with other
players, and a Terminal window for full access to FIBS and telnet. The user can
color code and keep private notes about other players (the color coding is also
used in the Chat window).
MacFIBS is freeware and is my contribution to the 'net.
The program requires a color Macintosh and MacTCP, and is available via ftp at
the Info-Mac Archives (sumex-aim.stanford.edu) as well as numerous mirror sites
around the world.
MacFIBS 2.0* is a self-extracting archive file: ftp://ftp.best.com/pub/fergy
--fergy
Paul Ferguson fergy@best.com
[MacFIBS is also available on AOL, as well as Info-Mac
mirror sites.
From: keithv@chiwaukum.CS.Berkeley.EDU (Keith Vetter)
Subject: TkFibs - an X interface to FIBS available
Date: 4 Apr 1994 19:31:41 GMT
Organization: University of California, Berkeley
Announcing TkFibs, an X based, graphical user interface to Fibs.
TkFibs is a tcl/tk client that provides a better interface to Fibs. It
displays two windows: one a graphical depiction of the board, the other
session window with Fibs ala the bottom two windows in tinyfugue.
I've been using the program for over 5 months now so it should be very
solid. I've run it on DecStations, SparcStations, HP 735 and Alphas.
The biggest caveat is that it requires TCL/TK to run. TCL/TK is a
very nice scripting / user interface package for X. It is available
from ftp.cs.berkeley.edu in /ucb/tcl.
TkFibs is located at ftp://shuksan.cs.berkeley.edu/pub/tkfibs
Send comments, suggestions, bugs, etc to:
tkfibs@shuksan.cs.berkeley.edu
Enjoy
keith vetter
FIBS/W is a Microsoft Windows(TM) based
client for
FIBS.
FIBS/W provides a graphical game board and mouse-driven interface to the FIBS
server. Most operations required to play a game can be executed using mouse,
keyboard, menus or toolbar buttons.
Powerful configuration options for many common Internet
host systems, and communications service providers, and a built-in
communications scripting language allow FIBS/W to automatically dial and connect
to FIBS via the Internet with a single mouse click.
To use FIBS/W you must be able to satisfy one of two
conditions:
- You must have modem access to a host computer
providing telnet services. If you can connect to the Internet using Windows
Terminal, you can probably use FIBS/W.
- You must have access to the internet via network
TCP/IP, PPP or SL/IP via the Winsock software interface. If you can find the
file WINSOCK.DLL on your system, you can probably use FIBS/W.
FIBS/W
requires version 3.1 or later of Microsoft Windows, or any version of Windows
for Workgroups or Windows/NT. FIBS/W will also run as a Windows application
under OS/2 2.X. FIBS/W does not currently support OS/2 Warp, or Netcom
Netcruiser accounts (although Netcom shell accounts do work).
FIBS/W is provided as Shareware. The registration fee is
US$40. This version of FIBS/W includes a Nag Screen (a mildly annoying dialog
which is displayed every time the program is run) but is otherwise fully
functional.
FIBS/W is available via anonymous FTP at: resudox.net in
the directory /pub/pc/windows/games/fibsw.
FIBS/W is available via the web at http://www.magic.com/~rdavies/fibsw.html
A WWW page for FIBS/W is available at ftp://resudox.net/pub/pc/windows/games/fibsw/html/fibsw.html
Best regards,
Robin Davies.
rdavies@fox.nstn.ns.ca
FIBS: Q
From: d9jesper@dtek.chalmers.se (Jesper Blommaskog)
"xibc" is an X11 client to the First Internet Backgammon Server (FIBS). It
is using the freeware packages Tcl, Tk and Expect. Normally, you have
to fetch and compile those to be able to run xibc, but not anymore
(provided you have a SunSparc!).
ftp.cd.chalmers.se:pub/xibc/xibc-X.XX.tar.Z
ftp.cd.chalmers.se:pub/xibc/xibc-X.XX.README
ftp://ftp.cd.chalmers.se/pub/xibc/
# If you need an executable (Sun-SparcOS 4.1.x only):
ftp.cd.chalmers.se:pub/xibc/binREADME
ftp.cd.chalmers.se:pub/xibc/BX.enc
# If you need an executable and don't have a Sparc, then you need
# to compile Tcl, Tk and Expect on your own. Here's the ftp addresses:
sprite.berkeley.edu:/tcl/tcl7.3.tar.Z
sprite.berkeley.edu:/tcl/tk3.6.tar.Z
sprite.berkeley.edu:/tcl/tk3.6p1.patch
ftp.cme.nist.gov:/pub/expect/alpha.tar.Z
(all but the patch are present at cme.nist.gov)
Features of xibc include:
* Log out from FIBS but keep the interface on the screen (maybe
iconified). Permits you to start xibc in your X startup files
and keep it up all time.
* X resource Tk*xibcLogfile that names a file where to log match
results. Example from my own file ~/.Xdefaults:
Tk*xibcLogfile: ~/spel/backgammon/xibc.log
* Resource Tk*xibcDelay that sets the programmed delayes in the
interface (the time interval between the different moves in a
sequence). Time is in milliseconds. May also be changed during
a session from a menu (but may not be saved).
Example: Tk*xibcDelay: 500
* The command line has some emacs/tcsh-style "cooked" line features:
C-n next line in history
C-p previous line in history
C-a first on line
C-e last on line
C-u delete whole line
C-d delete the character after the insertion marker
and also
C-s toggle "autoscroll" mode of the text window
C-l toggle logging of game to a log file
C-c log out from server and quit the interface
* There is a menu option called "Emergency". It may also be invoked
with "M-e". It reloads the whole board position from the server.
Nice when the interface screws up (it does sometimes).
* A menu option "Empty textwindow" will delete all rows in the text
window in access of 500. May someday become an X resource.
Don't try to push the interface to hard, that is, don't try to break
it. You will most definitely succeed (not hard at all, I guess). Since
I'm sort of an artist, I don't like breaking my own things so I
haven't really tried to find the bugs.
Special features (also called known bugs):
* Try the "look" command.
* Play a game against "You".
* Do several things at "the same time".
* Answer questions like "Accept double" or "join/leave" from the
command line, not by using the interface.
If you try them out, remember there's always the "Emergency" command
around.
-Jesper Blommaskog, author of "xibc"
-----
[Just a reminder to client writers. Some of us live behind 'firewalls',
or can only obtain phone access. Please remember us!]
Long Distance Backgammon. Play
backgammon by E-mail.
Program written by Perry R. Ross
(perry@aap.com)
From the ldb man page:
Ldb allows two people to play backgammon over a network
using electronic mail. It runs on character-oriented terminals, or emulators
thereof, using the curses screen package. It will run on most UNIX dialects, as
well as VAX-C under VMS 5.0 and above. Ldb handles all aspects of starting,
playing, checking, and scoring games. It enforces all normal game rules, as well
as several optional rules, and will not let you make an illegal move. When you
have made your moves, ldb will automatically package your move and send them to
your opponent.
The latest version is 1.3.2. Version 1.3 can be found in
directory volume36. Patches can be found in subsequent volumes at your favorite
comp.sources.misc archive site. Patch 1 is in volume 39. Patch 2 is in volume
41. Use: 'unix_prompt$ archie ldb' to locate the sources.
[from Perry]
I mentioned in
that patch that, for people who can't figure out how to get ldb or how to apply
patches, I'd be happy to send them a complete copy of the latest version. You
might want to put the same offer into the faq.
[Has anybody written a PC/Mac version using CC-mail via
a Novell network?]
[from Perry...]
Well, I'd
always intended to do a PC port, but just never got around to it. I was a bad
boy, 32-bit wise, so there would be a little effort involved making it 16-bit
clean. There's a package that simulates curses on a PC, I've heard. As far as
the particular mail transport, ldb doesn't really care. It puts outgoing
messages into a text file and executes a user-defined command to send the
message. Incoming mail can be read from a user-defined file (or pattern, to read
multiple files), which ought to be pretty transport-independent. It wouldn't be
that hard to port, I don't think.
[Anybody have a little ambition?]
The ldb 'game starter' operated by leopard@midnight.WPI.edu
(Leo Gestetner) has been shut down. [ Are there others? ...Mark]
LDB may be obtained on the net from: ftp://ftp.funet.fi/pub/vms/games/board/ldb.shr
and ftp://ftp.funet.fi/pub/unix/games/ldb.tar.gz
Garrett has shut down Netgammon as of late February
1995. It will be remembered and missed.
RSCARDS Backgammon on GEnie
GEnie (General Electric's Consumer Information Service)
offers on-line multi-player games in RSCARDS including backgammon.
Features:
- Graphic User Interface (GUI) is available for the
following computers: IBM, Atari ST, Apple IIgs, Macintosh Mono, Macintosh
Color, Amiga and Commodore 128 Mono
- TTY [text only] for non-supported formats.
- Friendly and fun atmosphere. Peak playing times are
evenings and weekends.
- Regular monthly prizes for eligible, high-score
players.
- Regular tournaments (see below).
- Game transcripts available immediately.
- Technical support and gaming discussion available in
the Multiplayer Games RoundTable.
- Access to other on-line multiplayer games including
RSCARDS chess, checkers, reversi, poker, blackjack and bridge.
How to
Access RSCARDS Backgammon on GEnie:
To sign up to GEnie, just follow these simple steps:
- Set your communications software for half duplex
(local echo), at 300, 1200, 2400 baud.
- Dial toll free: 1-800-638-8369, or in Canada,
1-800-387-8330. Upon connection, enter HHH
- At the U# prompt, enter JOINGENIE then press
<RETURN>
- When asked to enter a code enter: MMC524 [This will
waive your first $8.95 month subscription fee and give you an additional
$50.00 online credit during your first month on GEnie!]
- Have a major credit card ready. In the U.S. you may
also use your checking account number.
For additional information
including subscriptions and fees call: 1-800-638-9636
Once you have a GEnie account, simply type RSCARDS from
any GEnie prompt. This will take you to the main area, where you can download
GUI's and get general information on RSCARDS.
You can
reach the Backgammon page directly by typing M877. This will bring you to a menu
with specific backgammon information and access to play.
For questions and technical support, visit the Multiplayer
Games RoundTable. Type M1045 from any GEnie prompt and set to CATegory 29. There
are a variety of Backgammon, GUI and RSCARDS TOPics available.
TOURNAMENT INFORMATION:
GEnie Backgammon Tourneys are held quarterly. Sign-ups
start:
January 1st, April 1st, July 1st and October 1st
Tourney play starts the third Wednesday of those months
[Section on tourney rules is available online at GEnie]
[Note: GEnie is a service like compuserve... They have
a monthly fee and you pay by the hour for use. Contact GEnie for rates.]
Backgammon is among a package of eight
Macintosh only internet games offerred for a flat monthly fee ($9.95) by the
on-line service Outland, Inc. The games are advertised and offerred for ftp and
free trial at
http://www.outland.com/OutlandBackgammon.html Judging
from the web page it appears to share many features with FIBS plus providing a
nice built-in graphical (draggable pieces) interface.
Play By
E-Mail
While not quite in thte same league as FIBS, I have a
PBeM Server that supports Backgammon as one of its' games. Send mail to: pbmserv@vtsu.prc.com with 'help'
as the Subject: line for details, or visit Richards page at: http://coyote.vtsu.prc.com:8080/~pbmserv
This appears to be a backgammon server
that allows players with the hotjava browser to play other players. The server
has been created by Lee Smith.
http://www-leland.stanford.edu/~leesmith/JavaGammon.html
JavaGammon
http://www-leland.stanford.edu/~leesmith/hotjava.html
Info on Java extensions, programs, etc...
There are tournaments on
FIBS and
GEnie.
Tournaments on FIBS have been organized by David
Escoffery (davide), and David Eggert (snoopy).
contact: davide@sco.com
eggertd@aisb.ed.ac.uk
http://www.dai.ed.ac.uk/staff/personal_pages/eggertd/backgammon.html
GEnie has a quarterly single elimination tournament. $25
entry fee. Cash and credit prizes for first-fourth place. Hourly fee in effect
while you play. The draw is non random, in that previous winners are placed such
that they do not play each other in the first several rounds.
There are no backgammon servers other
than FIBS in operation on the internet at present.
Get 'OK.FAQ' from rec.games.bridge or
rtfm.mit.edu for info on the Internet bridge server.
Reported servers:
telnet okbridge@irc.nsysu.edu.tw 4321
login: okbridge password: okbridge
telnet bridge:bridge.0@zaphod.ttu.ee
login: bridge pasword: bridge.0
[The above servers have been reported to have an annual fee.]
telnet vanderbilt.okbridge.com
[Free guest trials are availble for the above server.]
A further source of information is available at:
http://www.cts.com/~okbridge/
telnet seabass.st.usm.edu 7777 or 134.53.14.112.7777
USA: telnet coolidge.harvard.edu 5555 or 128.103.28.15 5555
Sweden: telnet hippolytos.ud.chalmers.se 5555 or 129.16.79.39 5555
Taiwan: telnet 140.112.50.160 5555
A Xianqi Web page may be read at:
http://www.io.org/~sung/xq/xq.html
Othello(tm)/Reversi: telnet faust.uni-paderborn.de 5000
rafael.metiu.ucsb.edu 5000 128.111.246.2 5000
anemone.daimi.aau.dk 5000 130.225.18.58 5000
chess.lm.com 5000 129.15.10.21 5000
It has been reported that the chess servers now charge an anual fee.
igs.nuri.net 6969 203.255.112.3 6969
information: tweet@ig.nuri.net
tcasey@adobe.com
It has been mentioned that it is possible to play chinese chess on this
server.
Information about the checker playing program
``Chinook'' may be obtained by visiting the web page:
http://web.cs.ualberta.ca:80/~chinook/
Besides information, the Chinook page lets you play a
game on-line against the computer program which has beaten the human world
champion.
http://www.yahoo.com/Entertainment/Games
YAHOO WWW Games Directory
http://www.dcs.qmw.ac.uk/~steed/Games/bygame.html
Games and Puzzles on the Internet
http://www.io.com/games/servers.html
A list of game servers
http://www.inrete.it/games/telnet_e.html
A list of game servers.
Section C: ELECTRONIC BACKGAMMON: VS MACHINE
Ever since Gerry Tesauro finished
TD-Gammon, it was only a matter of time before a neural network program would
become available to the public. That time has come. Fredrik Dahl's masterpiece,
Jellyfish, is a breakthrough for backgammon. Both the checker play and cube
action of the program are at an expert level, making Jellyfish a truly enjoyable
and challenging competitor. In addition the program looks over your plays and
points out when you have made a serious error, making it extremely valuable for
learning purposes.
Jellyfish is run under Windows for the PC. Moves are
made with the mouse, and can be done very quickly and efficiently. The display
is nice and easy to see. Some additional features of the program:
Plays both single games and matches (yes, it understands
match equities). Allows the user to construct positions and save them. Gives the
user the program's evaluation of the equity of a position upon request, and the
evaluation function is surprisingly accurate. Tells the user when he has made an
error in checker play or cube decision, making the program the most valuable
tutor in the world.
In addition, a separate version is expected which will
also permit the user to roll out positions. In the past computer rollouts were
always suspect because the program didn't play well enough so the results could
be very distorted. This is no longer the case, since Jellyfish definitely plays
well enough to handle almost any position adequately. Results from its rollouts
can be trusted, and we will be able to find the answers to many backgammon
questions which we previously did not know.
For the casual player, Jellyfish provides an excellent
opponent and a way to improve while playing. For the serious student of the
game, this program is an absolute must. Our knowledge of the game is about to
take a quantum leap, and the player who does not have access to Jellyfish will
be left far behind.
Kit Woolsey
- JellyFish Tutor 1.2 for MS-Windows. US$ 110.
- JellyFish Analyzer 1.0 for MS-Windows US$ 220.
- The Analyzer, will in addition to the Tutor, contain
a rollout module It will be release Jan 16 1995. If you own the Tutor, the
Analyzer may be purchased for the difference in price. If you order the
Analyzer before Jan 16 1995, the Tutor will be shipped immediately, followed
by the Analyzer when ready.
- Order from:
- EFFECT Software A/S
P.O. Box
56 Skoyen
N-0212 OSLO
Norway.
Please use
International Postal Money Order, or Visa. If you use Visa, send the account
number, date of expiration, amount and signature. You may also send a check, but
in that case please add $10 for expenses.
Hardware requirements: 386sx or better
Software requirements: Windows 3.1
The JellyFish programs come on 3.5'' diskettes.
Also available from: The GAMMON PRESS , Carol Joy Cole and The Dansk Backgammon Forlag.
The program, and technical support are also available
from Larry Strommen within the USA. Contact:
L. A. Strommen; 6866 Meadow View Dr.;
Indianapolis, IN 46226
Tel: (317) 545-0224 E-mail: diceman@indy.net
Fredrik Dahl may be contacted at
fredrikd@ifi.uio.no
It should be noted that JellyFish uses a copy protection
scheme. The program requires that you ``confirm'' your installation once a
month, at the first of the month, by inserting the original disk. There are no
limits to how many machines you may install the program on. The DOS rollout
module is not protected at all, although the rollout files must be created using
JellyFish.
From wbitting@crl.com Sat Sep 16 20:40:09
PDT 1995
Article: 9552 of rec.games.backgammon
From: wbitting@crl.com (William C. Bitting)
Newsgroups: rec.games.backgammon
Subject: TD-Gammon & IBM Family FunPak
Date: 14 Sep 1995 22:41:22 -0700
Excerpted from:
OS/2 Warp
Monthly Newsletter September 95 (starting at p86
of
176, ascii version)
by Jeri Dube
(This section is out of sequence as presented in
original article.)
Although playing backgammon on a computer that plays as
well as a world class master seems somewhat awe-inspiring, you can work up to
it. The game comes with five skill settings, where each higher setting uses an
increasing larger and more complex neural network as its underlying engine. If
you want to use TD-Gammon to improve your backgammon skills, it is quite good as
a learning device. Not only do you get feedback from the results of your playing
but the system is quite supportive of you. It gives a modest, `I win' message
when you lose and a hearty `Congratulations, you win!', when the computer loses.
To embody this expert backgammon-playing neural network
into an OS/2 game, IBM Research hired Keith Weiner, a professional PC game
developer, to add a front end written for OS/2's presentation manager. TD-Gammon
is fully 32-bit and takes full advantage of OS/2 Warp's multi-threading
capabilities. Like all presentation manager programs, TD- Gammon comes with a
settings notebook where you can set things such as the background color and the
animation speed.
Given the success of the TD-Gammon game, I asked Gerry
what his next neural network game would be. He told me that researchers have
used other games such as Chess, Othello, and Go with varying degrees of success
to study neural network learning. None have been as successful as backgammon.
Gerry theorizes that the stochastic element of backgammon (i.e. throwing the
dice) is what makes backgammon so useful in modeling the self-learning process.
With that in mind, Gerry's next venture into self-learning is with financial
time series analysis. If that project is as successful at learning as the
backgammon game, then I'm really looking forward to that program.
For more information on Gerry's work, you may want to
read his article ``Temporal Difference Learning and TD-Gammon'' published in
Communications of the ACM, volume 38, number 3, pp. 58-68 (March 1995).
(The newsletter article starts here and ends with the
above 4 paragraphs.)
When most people think of IBM Research, they tend to
think of fractals, scanning- tunneling-electron microscopes, or high temperature
superconductivity. Games are not usually one of the thoughts that come to mind.
However, the TD- Gammon game included in the IBM Family FunPak for OS/2 Warp was
developed by IBM Research.
By virtue of being created at such an auspicious place,
you would think that this version of backgammon is quite special. Well, to be
quite honest and not so humble, it is! TD-Gammon is the most advanced computer
version of backgammon. It can play at the most advanced levels. If the system
were a human, it would be rated as a World Class Master.
TD-Gammon was developed by IBM Research Staff Member,
Gerry Tesauro. Gerry is not a game developer, rather he is a theoretical
physicist who has been working in the area of neural networks and artificial
intelligence for several years. He did not initially intend to develop an OS/2
game for the Family FunPak. All he wanted to develop was a basic research
project to study learning algorithms that would enable a computer to teach
itself a task.
Gerry chose backgammon as the task because it appeared
to be a good domain in which a neural network might work well. At this point you
may be wondering now that I've mentioned it twice, what is a neural network?
Well, in short, it's a model of interconnected neurons (also known as nodes)
that was inspired by the logical neurons in the human nervous system. Each
connection between neurons has a particular weight value associated with it.
In the case of backgammon, the state of the backgammon
board is fed into input neurons that have connections to hidden neurons (or
units). These hidden neurons in turn connect to an output layer that holds the
value of the state (that is, the chances of winning from that particular state).
The computation between the input neurons and the hidden neurons is a weighted
linear summation of all the input neurons. The result of the summation is put
through a thresholding function. This function compresses the value to lie
within a certain range of probabilities. (In case it ever comes up in
conversation, the function is known as a squashing function.) The squashing
function is a non- linear function. The non-linearity allows a system to learn
more complex functions.
To use this model to teach a system backgammon, all the
initial weights between the neurons are randomly set. The neural network starts
from the opening backgammon position and plays both sides until one of the sides
wins. The outcome of the game is used as a reward signal for reinforcement
learning. That is, the neural network takes the outcome of the game and adjusts
the weights accordingly. The adjustments improve the network's ability to
evaluate board states for subsequent plays of the game.
This learning process is repeated hundreds and thousands
of times. Using an RS/6000 computer, the learning actually took about two weeks.
Gerry and his colleagues were amazed at how well the neural network learned to
play backgammon. The system kept getting better and better until it reached the
world class master status. Actually, the neural network could improve its play
even more with further training and a larger network.
TD-Gammon is available on the new IBM Family FunPak for
OS/2. The FunPak may be purchased from Indulable Blue [add url] or from a number
of other mail order software houses.
From: jiml@teleport.com (Jim Little)
Newsgroups: comp.os.os2.games,rec.games.backgammon
Subject: TD-Gammon available for free download
Date: 9 Nov 1995 23:52:45 -0800
IBM has made TD-Gammon, their supposedly groundbreaking
neural network- based version of Backgammon, available for free download. It
seems to be part of an attempt to promote their IBM Family FunPack. You can get
it by surfing to http://www.austin.ibm.com/pspinfo/funtdgammon.htm and
following the "Read the license information" link. You will have to fill out a
form with your name, address, etc. (But nothing forces you to enter valid
information. ;) )
From their web page: "TD Gammon requires OS/2 2.1 or
higher, an Intel 386-SX or higher, with Advanced and Expert levels requiring a
486-DX 33MHz or higher, and a minimum of 6 meg of memory is recommended."
-Jim Little (jiml@teleport.com)
[md] The original article is available at: http://www.austin.ibm.com/psinfo/m4bakgam.htm
Newsgroups: rec.games.backgammon
From: tesauro@watson.ibm.com (Gerry Tesauro)
Subject: TD-Gammon paper available by FTP
Sender: Gerald Tesauro (tesauro@watson.ibm.com)
Date: Wed, 2 Jun 1993 18:06:35 GMT
Disclaimer: This posting represents the poster's views, not necessarily
those of IBM.
The following paper, which has been accepted for publication in Neural
Computation, has been placed in the neuroprose archive at Ohio State.
Instructions for retrieving the paper by anonymous ftp are appended below.
---------------------------------------------------------------
TD-Gammon, A Self-Teaching Backgammon Program,
Achieves Master-Level Play
Gerald Tesauro
IBM Thomas J. Watson Research Center
P. O. Box 704
Yorktown Heights, NY 10598
(tesauro@watson.ibm.com)
Abstract:
TD-Gammon is a neural network that is
able to teach itself to play backgammon solely by playing against itself and
learning from the results, based on the TD(lambda) reinforcement learning
algorithm (Sutton, 1988). Despite starting from random initial weights (and
hence random initial strategy), TD-Gammon achieves a surprisingly strong level
of play. With zero knowledge built in at the start of learning (i.e. given only
a ``raw'' description of the board state), the network learns to play at a
strong intermediate level. Furthermore, when a set of hand-crafted features is
added to the network's input representation, the result is a truly staggering
level of performance: the latest version of TD-Gammon is now estimated to play
at a strong master level that is extremely close to the world's best human
players.
---------------------------------------------------------------
FTP INSTRUCTIONS
unix% ftp archive.cis.ohio-state.edu (or 128.146.8.52)
Name: anonymous
Password: (use your e-mail address)
ftp> cd pub/neuroprose
ftp> binary
ftp> get tesauro.tdgammon.ps.Z
ftp> bye
unix% uncompress tesauro.tdgammon.ps
unix% lpr tesauro.tdgammon.ps
For a list of articles written by Gerry Tesauro, check
out: A List
of Backgammon Articles in Science and Business
Expert Backgammon 2.1 for PC: List of features.
Mark Damish E-Mail: damish@ll.mit.edu
Here is some initial information on Expert Backgammon
version 2.1 for the IBM PC.
- Intro:
- I remember November 92, the first time I walked into
a backgammon club to participate in a tourney. After playing on FIBS for a
month, I thought that I was already a decent player. Wrong. I won a match, and
lost a match. Afterwards, I played a few games for $1/point. I reached a
simple and common holding game position where I was doubled, and thought
surely it was worth 25%, and took. My opponent, being helpful to a newcomer,
pointed out that the position was only worth about 15%. Later, I was able to
verify his claim using Expert Backgammon, and had my first 'benchmark'
position. I have since used Expert Backgammon to benchmark many other simple
positions, as well as to play hundreds of games against it.
- What it is:
- Expert Backgammon, (EXBG), is a program which allows
you to play backgammon against the computer either in a `money' or
`tournament' format. It also allows you enter a position, and let the computer
`roll it out' --- that is, to let it play both sides many times, and show you
the results. Expert Backgammon is currently one of the stronger computer
program available commercially, and the game version is quite affordable!
- Brief Description of EXBG versions:
- Expert Backgammon is currently available in two
releases: 1.61, and 2.1. Release 2.1 has 3 different versions, with different
features
EXBG 2.1 GAME VERSION $50. Plays the game of backgammon.
EXBG 2.1 EXPERT VERSION $150. Plays BG, and Rolls out positions.
EXBG 2.1 PRO VERSION $300. Plays BG, and Rolls out positions. Has
some advanced rollout features.
EXBG 1.61 EXPERT VERSION $100 Plays BG, and Rolls out positions.
Upgrades from EXBG 1.61[expert] to EXBG 2.1[expert] $60
Upgrades from EXBG 1.61[expert] to EXBG 2.1[pro] $200
There are other upgrades available for the other versions as well.
- Some Random Features (pro version):
-
- Fast non mouse interface for moving the checkers.
- Match or Money play options.
- Optional Jacoby rule.
- Cube profile statistics.
- Save positions for future evaluation.
- Save games to be played back later.
- Computer can suggest a move.
- Shot counter.
- Computer can finish game, when it becomes routine.
- Woolsey or 35% (Friedman?) match equity table for
matches.
- Pip count.
- Quick or Extended cube searches. Speed vs.
accuracy.
- Rollouts:
- Random dice.
- Sequenced dice for one or two sides. That is all
36 possible starting combinations for one or two sides.
- Rollout multiple positions simultaneously in
batch mode.
- Duplicate dice when rolling out multiple
positions.
- Speed:
- Less than 5 seconds per game to play a game from the
starting position on a 486DX2-66 processer. This assumes that the graphical
display is disabled. ie: moves and rolls are not shown.
- Strength:
- Seems stronger than 1.61.
Strength is also a somewhat controversial subject. Can
the machine play a complex prop as well as a human who is familiar with the
position? Does the machine have any 'blind spots' in early game play? Can it
'work a prime' to get a second checker when needed? Does it blitz too often,
not enough? 'Seems stronger' is as far as I'll venture until I learn more
about how to play the game myself! See the section on How good is good? for Bill Roberties
ratings of backgammon programs.
- Copy Protection:
- The program is copy protected. The distribution disks
will allow you to install the program to two hard disks, as well as allowing
you to run the program from the floppy. You may 'uninstall' the program from
the hard disk, back to the floppy for installation on another disk/machine if
required. It is best to uninstall the program before using backup programs, or
reformatting your hard disk/partition, as you can loose your installation.
Early versions of 1.61 used a differant scheme. They
simply stopped running after a certain date. The authur supplies free updates
to those with the early protection scheme.
- Sample Cube Analysis Screen:
B L A C K C U B I N G A N A L Y S I S
C U B E Game won by - lost by Gammons Backgammons Equity
location size cube play cube play won lost won lost /Game
Center 1 47 3 22 8 4 9 0 0 0.108
Black 2 8 0 0 11 0 0 0 1 -0.600
Red 2 0 21 2 2 4 0 2 0 2.000
Black 4 1 0 0 1 0 0 0 0 0.000
Red 4 0 1 1 0 0 0 0 0 0.000
Black 8 0 0 0 1 0 0 0 0 -8.000
01-21-1994 The Jacoby Rule was NOT USED
Only Money Play with Extended Cube Search
MARK won 61.074% of the games played
Equity for MARK = 0.349 per game for 149 games
Cubeless equity for MARK = 0.228
Net equity when Black took a 2 cube = -0.870 per game
Net equity when Red took a 2 cube = 1.879 per game
[An above average performance in this set.]
- Help Screen:
F1 = MENU OF AVAILABLE OPTIONS - Professional Edition
A - Automatic Double Ctrl+A - Automatic Concession
B - Beaver Ctrl+B - Clear Board
C - Checker Setup Ctrl+C - Cube Setup
D - Double the Cube Ctrl+D - Manual Dice Entry
E - Extended Cube Search Ctrl+E - Match Equity
F - New Player Name Ctrl+F - File - Alternate Path
G - Game - Money or Match Play Ctrl+G - Delete Game
H - On-Line Help Service Ctrl+H - Using Option Defaults
I - Invert Position Ctrl+I - Alter the Starting Position
J - Jacoby Rule Ctrl+J - Title for Rollout
K - Monitor Type Ctrl+K - Player Cube Profile
L - Level of Difficulty Ctrl+L - Listing of Game
M - Take Back Move
N - Sound Ctrl+N - Start New Game
O - Options Currently Selected Ctrl+O - Open Position
P - Player on Roll Ctrl+P - Print Position
Q - Show PiP Count Ctrl+Q - Black Shotcounter
R - Replay Game Ctrl+R - Rollout Position
S - Speed of Checker Movement Ctrl+S - Save Position
T - Suggest Move for Black Ctrl+T - Match Win % Table
U - Skip Save Game/Position Ctrl+U - Delete Position
V - Black Detailed Shotcounter
W - Expert to Finish the Game Ctrl+W - Write Rollout to Disk
X - Cancel Move and Reroll Dice Ctrl+X - Print Disk Rollout
Z - Zero the Score
Spacebar - Roll the Dice Escape - Leave EXBG
- Rollout summary printout:
EXPERT BACKGAMMON ROLLOUT SUMMARY
Date: 12-12-1993 Games viewed = 0
File: 65_21_A Summary only = 1296
Version: 2.1 1296 games rolled out at
Used Hrs:Min:Sec 1:37:41 4.522 seconds per game.
Batched Duplicate Rollout of 1296 games.
No doubling allowed - Cube at 1 level
[ Diagram of board position was cut from here...]
O on roll.
O won
45.83% games 594
11.42% gammons 148
0.46% backgammons 6
---------------------------
57.72% of the games 748
+0.170 points per game.
X won
32.18% games 417
9.41% gammons 122
0.69% backgammons 9
---------------------------
42.28% of the games. 548
-0.170 points per game.
The dice rolls were generated randomly.
Note: There is also a screen which shows the results of several games
rolled out simultaniously.
Note: Above screens were 'captured' by redirecting my printer port to
a file, and have been edited slightly.
- Ordering Info:
- Tom Weaver
Expert Backgammon
8063 Meadow Road, # 108
Dallas, Texas
75231
Call: Expert Backgammon (214) 692-1234 M-F 10am-10pm
Central US time. An answering service answers when Tom isn't around.
Tom also has an email address: tomweave@netcom.com
Also available from Carol Joy Cole and The
GAMMON PRESS.
A Macintosh version of Expert Backgammon is also
available. Contact the sources above, or Tom Johnson (auther) directly atkomodo@netcom.com. A demo of the
Macintosh version is available for anonymous ftp from: ftp://ftp.cybercom.net/pub/users/damish/backgammon.
A
backgammon program for MS-DOS
"I believe Blot makes primarily other mistakes than the
backgammon programs I know. Due to the selective (rather speculative) style
definitely Blot is tactically the weakest program of all (with outrageous
blunders in the endgame), but positionally not as flawed as many computer
opponents."
Blot has good results against many backgammon progs (see
blot.doc).
If you have any comments on blot, the authors are glad
to receive mail from you. In case you don't want to bother with sending (real)
mail, you can E-mail me, I'll forward your comments to the authors.
The programm is still being developed, so be prepaired
to get new blot versions soon :-).
I hope you enjoy the programm!
Alexander Fuchs
available from: ftp://ftp.cybercom.net/pub/users/damish/backgammon/
Backgammon, by George!
Version 1.50 has been available since June 1995. This shareware ($15) program is
suitable for Windows 3.x and Windows 95 and may be downloaded from the WINFUN
forum in CompuServe and other places on the net. If you can't find it contact
the author:
George Sutty P.O.Box 6247
Huntinton Beach, CA 92615
USA
sutty@ix.netcom.com
Backgammon for Windows version 0.6
A fairly weak backgammon program for windows, originally
introduced in 1990.
bg06 is available for anonymous ftp from
ftp.cica.indiana.edu in the directory /pub/win3/games/
There is a backgammon game for the Macintosh by Stephen
Young, Debra Willrett, and David Young. The 1.0 version is fairly widespread,
although there is a 2.0 version (dated May 25, 1989) available on America
Online. The play is pretty weak, and the graphics are designed for the original
small B&W Mac screen, but if you're really bored and can't find a human
opponent, it works. The game is freeware, and you get what you pay for.
-- Paul Ferguson
For those interested in a good bg game for Windows, why not
try Death By Backgammon for Windows. It runs under Windows 3.1 or OS2/2.1, and
features animated dice/game pieces, comprehensive help, move undo, suggest move,
speed control over all motion, and is fully resizeable. Best of all, its
strategy is very competitive (I wrote the thing, and have a roughly 50:50
average against it over many hundreds of games). If you are interested, I would
be happy to send out a shareware version for evaluation. The shareware version
is fully functional, except that the computer's moves are painfully slow. There
is also a DOS version, which is a bit older, but still features animation and
VGA graphics.
To get hold of the program there are two options:
- To get the shareware version, (free of charge) let me
know and I can email you a zipped uuencoded version of the shareware.
Alternatively, I can send you a floppy if you provide your mailing address.
- To get a registered copy of the program, send a
cheque made out to Chris Kanaris, and I will email or post as above.
Prices: Shareware - Nil
Registered: WIN $AS 35.00, $US 30.00.
DOS $AS 20.00, $US1 5.00.
My Postal address is: Chris Kanaris
PO Box 495
Essendon,
Victoria, 3040
Australia.
[ People who have tried the shareware version claim that
the program plays a pretty weak game. Hopefully computer bg game writers will
continue to make their games stronger and stronger. Not and easy task! ]
From: klasen@obelix.uni-muenster.de at SMTP-Post-Office
Subject: contribution to FAQ
xgammon.0.96
xgammon is a BG-playing programm originally written for Linux.
The authors are Lambert Klasen (klasen@uni-muenster.de) and
Detlef Steuer (steuer@amadeus.statistik.uni-dortmund.de, blotstorm on FIBS)
Features:
xgammon is Freeware under the Gnu Copyright.
xgammon has nice graphical interface, you move by clicking on mousebuttons.
You can do a maildump for a position, that means you get a file with
a FIBS style board with current position in it.
You can do money game or tournament game.
You can edit positions via mouse on the board or via ascii plain text files.
You can turn doubling on and off.
You can do rollouts with doubling turned on and off.
The programm uses an endgame database for perfect bearing off (and for
reasonable running game).
There is an compi_finish for shorten the boring part of the game.
and and and ....
You can have fun with xgammon.
Give xgammon a chance getting compiled on your machine.
We'd like any reactions, especially porting reports to OS different
from Linux.
(heard of AIX and Sun OS compiling), especially bug reports .
Send a mail if you use it, please!
You find the latest Version of xgammon, at the time xgammon.0.96.tar.gz,
at ftp sunsite.unc.edu /pub/Linux/X11/games/strategy/xgammon.0.96.tar.gz.
Detlef Steuer
steuer@amadeus.statistik.uni-dortmund.de
(blotstorm on FIBS)
Other programs that can be found around the net include: backga.zip,
egagam20.zip pcgame.zip, pcgam416.zip and pcgammon.zip. Check DOS related ftp
sites for locations.
Program Name Source Type Score
----------------- ------ ------ ------
TD-GAMMON 2.1 N/A N/A -0.05
TD-GAMMON IBM OS/2
JellyFish 1.0 Dahl IBM-PC/Win -0.15
Expert BG 2.1 Weaver IBM-PC -0.20
Expert BG 1.61 Weaver IBM-PC -0.35
Championship BG Spinnaker IBM-PC -0.66
Expert BG Komodo Macintosh -0.82
Sensory BG 2 Scitek Portable -0.94
Backgammon Odesta IBM-PC -1.20
BG by George GS Labs IBM-PC/Win -1.52
Video Gammon Baudville IBM-PC -1.61
PC-Gammon Repsted IBM-PC -3.67
Gammon Gakken Portable -12.40
Windows BG Baudville IBM-PC/Win -13.83
Gammon Pal Fidelity Portable -15.63
Micro BG Fidelity Portable -15.53
Games People Play Toolworks IBM-PC -26.60
[ From the 1994 The GAMMON PRESS catalog.
and program reviews in Inside Backgammon.]
Score is the number of points won per game, on average, against a top
flight human player. Very large numbers are caused by bad doubling algorithms
which cause a program to double when behind (typically when primed but ahead in
the race), causing the computer to lose some very large cubes.
There are two basic ways that a
computer can play a game as well as or better than humans. One is to be really
smart, the other is to do an awful lot of work. The general strategy most
game-playing programs use is to use an evaluation function that isn't very
smart, but to make up for it by looking ahead a lot of moves (doing a lot of
work).
With chess, there are typically 20-30 moves by each
player per turn. With backgammon, there are 21 unique rolls and often 4-6 ways
to play each one (not counting doubles with could have 10 or more ways of
playing). This makes it very difficult to look ahead very many levels. Looking
ahead 3 moves by both players examining all possibilities when there are 25
choices at each play requires evaluating "only" 244 million positions. If there
are 90 ways to play each move, there are 530,000 million, positions.
With a game like chess, one can discard all but the best
5 or 10 plays per person. With backgammon, there are always 21 different choices
of best plays, depending on the dice. This makes it crucial to have an excellent
evaluation function.
The difficulty in doing this is that factors such as the
race have a different effect on the value of the position depending on what
stage the game is in. Consider the concept of timing -- hard enough for people
to grasp, extremely difficult for computers.
Another example of the difficulty of evaluating plays:
It's almost always beneficial to close out your opponents checkers. But if
you've hit one checker and you almost have to hit a second to be able to win,
closing out your opponent is very bad.
-michael j zehr
Further information relating to machine learning in
games may be found at: http://forum.swarthmore.edu/~jay/learn-game/systems/gammon.html
[NOTE: Programs that play backgammon are in the section:
Are there
any BG programs out there for my computer? Where are they? " ie: JellyFish
& Expert Backgammon.]
Boinq is a program which analyzes bearoffs.
It can handle any position where both sides have all their men in the inner
board. All results are cubeless. For any position you enter, you can get the
probabality of each side winning, equity on a 1-cube, proper way to play any
roll of the dice, and a distribution of probabality of bearing off in any number
of rolls. The results are displayed virtually instantaneously, since they are
read directly from a data file rather than done by simulation. Very user
friendly and easy to use -- I use it a lot for a quick check on bearoff
problems. Program takes about 4 meg, so have some room on your hard disk.
Produced by Hal Heinrich -- cost is $100 I think. Can contact Hal at:
Hal Heinrich
#203, 215 14th
Ave. SW
Calgary, AB
Canada T2R
0M2
Phone: (403) 234-9944
E-mail
address: heinrich@cuug.ab.ca
Kit
[Note: Also available from Carol Joy Cole]
[Note: As far as I know, this is a PC program ...Mark]
Hyper-Backgammon is a short
version of backgammon where each side has three checkers. In initial position,
these checkers start on opponent's ace, two and three points. From then on
normal backgammon rules apply. Cube is in play, Jacoby rule, gammons and
backgammons count (and are quite frequent). Proper play is much more subtle than
might be imagined. The program produced by Hugh Sconyers plays the game
perfectly, since Hugh has established a full data base which has the equities
for all possible positions, and the program will tell you if you make an error.
It can be a lot of fun to play, the games go quickly, and by playing the program
you learn pretty quickly the correct strategies.
Kit
[Note: Also available from Carol Joy Cole and The
GAMMON PRESS] [Note: The distribution is a CD-ROM for the PC ...Mark]
Volume 1: Bearoff Equities and Backgame Probabilities
INTRODUCTION:
This CD-ROM disk contains two large databases of
equities and probabilities. One database has the exact equities for all bearoff
positions when each side has 9 men or less. The other database contains the
probabilities for getting hit or hitting a man when one side has a backgame of 4
men or less.
The bearoff database is over 400 MB and contains the
answers to over 100,000,000 bearoff problems! In addition, you can do a MONTE
CARLO simulation on any bearoff position with more than 9 men on a side. The
program also has a feature which finds the best move given a specific position
and dice roll.
The backgame database is 90 MB and contains the answers
to over 22,000,000 backgame positions (some of them are not legal positions).
There is a feature which finds the best move given a dice roll and position.
These databases will help settle many questions about
fair settlement and the best moves.
BEAROFF:
Once the program has been loaded you can enter any
bearoff position, assuming that all men are in their home board. If both
positions you enter have 9 men or less, the program will give you the EXACT
equities for all 4 cube positions - NO CUBE, ROLLER'S CUBE, CENTER CUBE, AND NON
ROLLER'S CUBE. In addition, it will display the proper cube decision. These cube
equities, as throughout this bearoff program, are the equities if you roll with
the cube in that position.
To compute the proper settlement in any bearoff
position(assuming you are going to roll with the cube in that position) you
would multiply the equity times the value of the cube.
After the equities are displayed, you are given a chance
to find the best move for the position you have just entered. Keep in mind your
best move in the bearoff is the one that leaves your opponent with the smallest
equity. The best move will often be different depending on the cube position.
If you enter a bearoff position where one or both side
have more than 9 men you will enter the MONTE CARLO subroutine. You will first
be asked how many simulations you want to run. There is a limit of 30,000. This
Monte Carlo subroutine rolls the dice and moves the men until each side has 9
men or less. Then it looks up the exact answers from the database. With a few
hundred simulations the NO CUBE result should be very close to exact. The other
cube positions results will be distorted by the fact that this subroutine
assumes that there is no doubling until the positions are back in the database
(ie each side has 9 men or less). For example, if you have 12 men and the cube
is in the center and your opponent has 12 men, the cube will stay in the center
until each side has 9 men or less. At that point, the subroutine will retrieve
the exact equity from the database.
BACKGAME:
Once the program has finished loading you can enter any
backgame position provided the position not bearing off has 4 men or less. The
side bearing off can have from 1 to 15 men. The program will return two
probabilities: one for each side being on roll. These results are the
probabilities that the side bearing off will have a man hit.
After the probabilities are displayed, you are given a
chance to find the best move for the position you have just entered. Keep in
mind that the best move for the position bearing off is the one that leaves the
other side with the smallest probability of hitting a man. The best move for the
side in the backgame is the one that gives it the highest probability of hitting
a man.
This program assumes that the side playing the backgame
has infinite timing; ie, he will never be forced to leave his opponent's home
board. It follows from this assumption that the backgame side can take some,
part or none of any roll.
Sony's MMCD Player:
This disc works on an MMCD player also. Sony makes this
very small CD-ROM player which takes special exe files. This disc works both for
dos and MMCD. The MMCD version is the same as the PC version except in a few
places. To start the program place the disc in the MMCD player and turn the
power on. If you plan to use the bestmove feature and the MONTE CARLO routine
you will need to put the Volume #1 disc in the player after the program loads.
The MONTE CARLO simulations are limited to 20,000 games.
Volume 2: Bearoff Equities for 4 Points and 15 Men
INTRODUCTION:
This CD-ROM disk contains two databases of equities for
the bearoff. The first database(4X15) has the exact equities for all bearoff
positions when each side has 15 men or less on the first 4 points. The second
database(3X15) contains the exact equities for all bearoff positions when each
side has 15 men on the first 3 points. The second database is a subset of the
first. The smaller database(3X15) is included for a number of reasons, which
will be explained later. Everything that follows applies to the 3X15 database
when the 4's are change to 3's etc.
The bearoff database for 4 points and 15 men is over 240
MB and contains the answers to over 60,000,000 bearoff problems! In addition,
you can do a MONTE CARLO simulation for any bearoff position where one or both
sides have men on the 5 or 6 points(4,5 or 6 in the case of the 3X15 database).
The program also has a feature which finds the best move given a specific
position and dice roll.
These databases can help settle many questions about
fair settlement and the best moves.
BEAROFF EQUITIES:
Once the program has loaded you can enter any bearoff
position, assuming that all men are in their home board. If both positions you
enter have all their men on the first 4 points, the program will give you the
EXACT equities for all 4 cube positions - NO CUBE, ROLLER'S CUBE, CENTER CUBE,
and NON ROLLER'S CUBE. In addition, it will display the proper cube decision and
probability for winning in the no cube case. These cube equities, as throughout
this bearoff program, are the equities if you roll with the cube in that
position. If you find an equity greater than 1.000 or less than -1.000 this is
due to the fact that a position with 15 men can still lose a gammon.
To compute the proper settlement in any bearoff
position(assuming you are going to roll with the cube in that position) you
would multiply the equity times the value of the cube.
After the equities are displayed, you have an
opportunity to find the best move for the position you have just entered. Keep
in mind your best move in the bearoff is the one that leaves your opponent with
the smallest equity. The best move will often be different depending on the cube
position.
If you enter a bearoff position where one or both side
have men on the 5 or 6 points you will automatically enter the MONTE CARLO
subroutine. You will first be asked how many simulations you want to run. There
is a limit of 30,000. This Monte Carlo subroutine rolls the dice and moves the
men until each side has all men on the first 4 points. Then it looks up the
exact answers from the database. With a few hundred simulations the NO CUBE
result should be very close to exact. The other cube positions results will be
distorted by the fact that this subroutine assumes that there is no doubling
until the positions are back in the database (ie each side has all men on the
first 4 points). For example, if you have 12 men on the one and 3 men on the 5
point versus the same, the cube stays in the current position until both sides
have all their men on the first four points. At that point, the subroutine will
retrieve the exact equity from the database.
THE 3X15 DATABASE:
There are several reasons that this database is
included. First, it is small(only 10 MB!). This will allow it, if you desire, to
be copied to your hard disk(be sure to copy BEQT4X15.EXE, helvb.fon and
tmsrb.fon). Secondly, the 3X15 database has an advantage in speed when doing
MONTE CARLO simulations. Simulations will run faster at the expense of some
accuracy. If you have copied the 3X15 database to your hard disk it will run
MONTE CARLO simulations substantially faster.
Volume 3: Bearoff Equities for 6 Points and 10 Men
INTRODUCTION:
This CD-ROM disk contains a database of equities for the
bearoff. The database has the exact equities for all bearoff positions where one
side has 10 men in the home board and the other side has 10 men or less in the
home board.
The bearoff database for 6 points and 10 men is over 625
MB and contains the answers to over 156,000,000 bearoff problems! In addition,
you can do a MONTE CARLO simulation for any bearoff position where one or both
sides have more than 10 men. The results of these simulations will be more
accurate than the results from Volume #1. The program also has a feature which
finds the best move given a specific position and dice roll. You will need
Volume #1 to use the MONTE CARLO feature and you may need Volume #1 for the best
move feature.
These databases can help settle many questions about
fair settlement and the best moves.
BEAROFF EQUITIES:
Once the program has loaded you can enter any bearoff
position, assuming that all men are in their home board. If you enter a position
where both positions have 9 men or less you will get an error message because
all these positions are on Volume #1. For positions where one side has 10 men
and the other side has 10 men or less, the program will give you the EXACT
equities for all 4 cube positions - NO CUBE, ROLLER'S CUBE, CENTER CUBE, and NON
ROLLER'S CUBE. In addition, it will display the proper cube decision and
probability for winning in the no cube case. These cube equities, as throughout
this bearoff program, are the equities if you roll with the cube in that
position.
To compute the proper settlement in any bearoff position
(assuming you are going to roll with the cube in that position) you would
multiply the equity times the value of the cube.
After the equities are displayed, you have an
opportunity to find the best move for the position you have just entered. Keep
in mind your best move in the bearoff is the one that leaves your opponent with
the smallest equity. The best move will often be different depending on the cube
position. For some positions you will need Volume #1.
If you enter a bearoff position where one or both sides
have more than 10 men you will automatically enter the MONTE CARLO subroutine.
You will need Volume #1 to use this subroutine. You will first be asked how many
simulations you want to run. There is a limit of 30,000. This Monte Carlo
subroutine rolls the dice and moves the men until both sides have 10 men on
less. Then it looks up the exact answers from the database. With a few hundred
simulations the NO CUBE result should be very close to exact. The other cube
positions results will be distorted by the fact that this subroutine assumes
that there is no doubling until the positions are back in the database (ie each
side has all men on the first 4 points). For example, if you have 12 men on the
six point and 3 men on the 5 point versus the same, the cube stays in the
current position until both sides have 10 men or less. At that point, the
subroutine stores that position and later retrieves the exact equity from the
database on Volume #3 or Volume #1.
Sony's MMCD Player:
This disc works on an MMCD player also. Sony makes this
very small CD-ROM player which takes special exe files. This disc works both for
dos and MMCD. The MMCD version is the same as the PC version except in a few
places. To start the program place the disc in the MMCD player and turn the
power on. If you plan to use the bestmove feature and the MONTE CARLO routine
you will need to put the Volume #1 disc in the player after the program loads.
The MONTE CARLO simulations are limited to 20,000 games.
Available from Carol Joy Cole, The GAMMON PRESS and The Backgammon Shop for $99 per volume.
If you have any comments or questions, please forward
them to the
Hugh Sconyers
sconyers@bga.com
With the MatchQiz software, Kit Woolsey (long time contributer to
Inside Backgammon and Backgammon with the Champions, author of How to Play
Tournament Backgammon, and currently ranked #9 in the world) has added his name
to the short list of backgammon indispensables. The very short list. Magriel,
Robertie, Woolsey. I think that's all you really need.
MatchQiz is more than very good; it is the single best
tool I know of for transforming your game from intermediate to expert. First let
me describe the format:
You choose a match from a menu, and the computer shows
you the starting position and opening roll. *Then you choose your play.* Now you
get to see the actual play, plus Kit Woolsey's commentary. This happens for
every play, every cube decision. It is an improvement over printed annotated
matches in several ways:
Convenience and speed. How many times have you followed
a match on your own board only to find that the moves and commentary have
stopped making sense? Perhaps you moved the wrong piece two rolls ago? Or was it
three rolls ago? Maybe you should just start this game over. Never again with
MQ.
How many hints do you receive when going over printed
matches? You can see that the player drops the cube because a new game begins
next page. Or did the index card you use to cover the bottom of the page slip,
denying you the chance to come up with your play without seeing Svobodny's? Not
with MQ.
Perfect use of default options. Do you have a tendency
to forget the cube in complex positions? MQ will let you make that mistake --
but will chide you for it.
Of course the format wouldn't mean much if the
annotations weren't up to par. But they are exquisite. Woolsey is complete,
concise, and entertaining. He covers all aspects of play, from the most
elementary opening moves to the most subtle match equity considerations. Here is
one small sample:
``This is an expert play which many players
would not find. If Magriel quietly plays 13/5, O'Laughlin will be free to make
any point that his dice dictate, and Magriel will be poorly placed in the
upcoming prime vs. prime battle since he will have two men back will
O'Laughlin will have only one man back. Magriel's play forces O'Laughlin to
attack on the bar point whether he wants to or not, thus preventing him from
making optimal use of such point making numbers as 42 or 51. It is thematic
when you have the better board and your opponent has one man back to split
your runners to make it difficult for your opponent to catch up in the
board-building battle.'' You get this level of analysis after every nearly
every play.
One final benefit that might go unnoticed is volume.
Woolsey has 18 matches available now, and volume four is due soon. That is quite
a library. Often an annotater will mention some general theme or principle and
show how it applies it to a specific position. But would that principle apply if
the position were slightly different? With the MQ library, you will be able to
compare similar positions and examine if the same principles apply. This is
especially true for the opening phase of the game, where the same sorts of
decisions come up all the time.
Woolsey sells one match for $20, or a set of 6 for $100.
This is a fantastic deal considering that most printed annotated matches go for
$20 or so, and I guarantee that you'll get more use out of these. Write to:
Hal Heinrich
#203, 215 14th
Ave. SW
Calgary, AB
Canada T2R
0M2
or call (403) 234-9944
Jeremy Bagai
[Note: Also available from Carol Joy Cole and The
GAMMON PRESS] [Note: This program is written for the IBM-PC, but it 'should'
run under "Soft-PC" for the Macintosh, any level, as the graphics are CGA
(640x200x2) ...Mark]
From: kwoolsey@netcom.com (Kit Woolsey)
Subject: Matchqiz Demo
X-Newsreader: TIN [version 1.2 PL1]
As many of you know, Hal Heinrich and I have produced
several annotated matches which run on computer (PC only -- sorry, you MAC
users). Hal is the programmer, I wrote the annotations. We have a demo match
which we give away for free, so I thought it would be a good idea to make it
available on the net both for the enjoyment of R.G.B. readers and, of course, to
entice you to purchase the other annotated matches. This is my first effort at
doing this so I may have screwed up somewhere -- if so, please let me know what
went wrong.
The program can be gotten via an anonymous ftp to
ftp.netcom.com. go to the directory /pub/kw/kwoolsey/gammon/matchqiz -- there
you will find 2 files. One is: readme, which is instructions for retrieving and
running the program. The other is: mqizdemo.exe, which is the program and
necessary files.
Feel free to distribute this demo package to any friends
who may be interested. If you run into any problems, please contact me at: kwoolsey@netcom.com
Kit Woolsey
[9507]
BPA is a program which has three basic features: The CPW
of a position, the correct way to play either one or two numbers of a roll, and
the number of rolls that bear-off all checkers in N turns.
The data base is uniquely compressed to fit within 20
megabytes and covers all home board positions, many bear-in positions where you
are setting up to bear-off, and a single checker race. In the single checker
race, any number of checkers can be off and the single checker can be on any
point from the 7 to the 19 point. Contact is allowed and a cpw will be
calculated if the opponents position is in the data base, but contact is ignored
in the calculation. Thus there are over 25 BILLION positions covered, not
counting all the single checker positions. That's a very efficient use of data
in memory. This concentration is possible because the answer for the cpw is an
approximation, not an exact figure. However, the approximation is very good.
Using an extended calculation from where others had left off, the expected error
in the range of 20 to 80 % is 0.04%. This is good enough to make all checker
plays accurately. ( Ok, you'll be able to find a few positions where the play
made is off by a very small fraction of 1%.)
When you have either one or two numbers you want to
play, BPA will give an answer very quickly by looking into the data base and
checking all possible ways to play the number(s).
The number of rolls to bear-off in N turns is very
helpful in understanding what is going on in certain positions, but is not for
everyone.
BPA is supplied with all the data calculated, and is
distributed on 14 diskettes. It is currently not copy protected but because of
the extent of illegal distribution that has gone on, it will be protected in the
future. Each program is individualized by showing a subsciber's name on the
screen. btw, if anyone has an illegal copy and wants to get registered at no
cost to them, they should contact me at diceman@indy.net for the procedure. This
is a limited time offer and will end when and if BPA is integrated into any of
the commercial game playing programs. Registered users will receive an upgraded
EXE file to be able to use BPA with such a program.
Larry Strommen diceman@indy.net
Available from Larry Strommen and Carol Joy Cole
Software Review: BG-SCRIBE, A Program By
Walter Trice.
Mark Damish (E-mail: damish@ll.mit.edu)
BG-Scribe -- A system for editing, replaying, and
printing backgammon matches for the IBM-PC by Walter Trice.
This is a program which I consider essential for myself.
It will:
- Allow one to enter matches from books and magazines,
or from those personally recorded, to be played back and studied later. Entry
is done using the numeric keypad. After a while, you end up being able to
enter a match very quickly, and become quite proficient at touch-typing the
keypad to boot. An `AT' style keyboard is preferable for entering matches.
- Play back matches purchased from Walter. Watching
great matches between the masters is definitely enlightening. My favorite
matches are the ones from annotated books. It shouldn't be hard to write a
program to convert a match from one format to another, if already have a
collection of matches. I've converted matches posted to Internet by Butch
Meese using nothing more than simple editor macros. The matches are currently
stored as plain ASCII files---one directory per match, and one file per game.
- Print out matches to disk or printer. I like to:
print out to disk, annotate my comments using a simple editor, then print the
resulting file to a printer. The program will also embed diagrams of positions
to the printout, but it uses IBM graphics characters, which might have to be
changed if your printer doesn't support them. The diagram feature is
especially useful for diagramming doubling decisions.
The program is run from one of two screens. The first is
a text screen with options for creating a new match, loading, saving, etc.
The second screen displays the backgammon board using a
CGA 4 color, 40 column text mode. Trust me here---this mode, with its X's and
O's, looks a lot better than backgammon boards I've seen drawn using the CGA 2
color graphics mode. Why CGA text mode? Probably because it will run on any
portable or palmtop machine, and likely on most PC emulators on other platforms.
It also runs fine in a window under MS Windows, without having to mess around
with a .pif file.
The second screen is where matches are entered or played
back. When playing back a match, you may see the dice, then the players choice,
and then see the move when the screen is updated. It is possible to go forwards
or backwards in a game. Although you may have entered your rolls in `landing
spot' format, the program can optionally display them in `from/to' format. It
only prints in the format which it was entered though. You may also mark
positions which you would like to have diagrammed.
As stated above, the user interface consists of two
screens. Unlike modern X/Windows/Mac programs which do a lot for you, this
program assumes that what you tell, or don't tell the program is exactly what
you want. I.E.: You can enter data, then quit the program. If you didn't save,
the program won't prompt you. It is like driving a standard after getting used
to cruise control. The program does a lot, you have to remember to tell it what
to do though! The learning curve has a slight incline, but the program is well
worth the time it takes to become familiar with it. I found the instructions
clear.
I want to start bringing pen and steno paper to local
events to record some `master games' for later entry/playback. I also want to
review some of my `bad games', searching for weakness. There is a lot to be
learned!
The price of the program is $50, including 11 matches.
Additional matches are available from Walter.
I am not affiliated with BG-SCRIBE in any way, except
for being a very satisfied customer. Please mention where you saw this article
if you should contact Walter. I did mention to him that I was going to write a
review---Last spring!
...Mark
- Contact:
- Walter G. Trice
549
Wachusett St.
Holden, MA 01520
(508) 829-3283
e-mail wgt@world.std.com
UPDATE
9512:
Walter as added a program which will convert FIBS
oldmoves format to BG-Scribe. I haven't tested it with output from rfibs.
Tom Johnson (
komodo@netcom.com) has written a
progam which is best described as a match equity calculator. Given: Match
length, Score, Gammon chances, Cube value, and wether the cube is on the last
roll, last two rolls, or a normal cube, the program will will calculate the
information shown in the sample screen shown below.
BLACK WHITE Menu
Input data --------------------- -----
match length (2 - 25) | 3 | 3 | L) Match length
score (0 - 2) | 0 | 0 | S) Score
cube value | 1 | | C) Cube value
chance of winning the game | 35 % | 65 % | W) Winning chance
chance of winning a gammon | 20 % | 20 % | G) Gammon chance
cube situation (1 - 3) | 3 | 3 | V) Cube situation
--------------------- N) New table
Static match winning chance with --------------------- X) Exit
no double this game | 45.99% | 54.01% | A) About demo
double and take | 41.00% | 59.00% |
double and drop | 60.44% | 39.56% | Cube situations
double, take and redouble | 35.00% | 65.00% | ---------------
Dynamic data (game win %) --------------------- 1) Last roll
minimum give point | 50.00% | N/A | 2) Last two rolls
optimum give point | 61.63% | N/A | 3) Normal cube
minimum take point | N/A | 30.43% |
doubling window (size) | 19.57% | 19.57% | N/A = Not applicable.
---------------------
Strategy: Black should not double.
His position is not strong enough.
THE MATCH STRATEGIST DEMO - copyright(c) 1994 Komodo Software
The demo program is limited to 3-point matches and is available for
anonymous ftp from:
machine: ftp.cybercom.net
directry: pub/users/damish/backgammon
file: matstrat.zip
The program runs on a MS-DOS machine, and the file needs to be 'unzipped'
with pkunzip 2.04 or later.
A commercial version, which gives equities up to a 25-point match may be
obtained from the auther for $45 + S&H.
[Now (3-95) advertized by Carol Joy Cole for $25.]
Newsgroups: rec.games.backgammon
From: spitz@irb.uni-hannover.de (Jan Spitzkowsky)
Subject: Recording and replaying games played on FIBS
Date: Wed, 31 Aug 1994 10:00:21 GMT
Hello fibsters,
I wrote two utilities for FIBS and i want to distribute them. The tools are
able to record (rfibs) own games and watched games and to replay (sfibs)
them.
I observe a lot of discussions about interesting positions in this group.
With 'rfibs' and 'sfibs' it is possible to extract special positions played
on FIBS (or even a whole game) and to comment it.
I am interested in a collection of good and interesting games, commented or
not, to improve my own playing. If someone wants to get these utilities
for collecting and sharing some games, too, the address of our ftp-server
is:
ftp.irb.uni-hannover.de
File: pub/spitz/bg/bg.tar.gz
Below I give a short description of the two tools:
Hope to find much interest for my work and many good games,
Jan (hotspot on FIBS)
Version 26th August 94:
Any bug reports, comments ... etc.
rfibs [<record file>] [-c]
'rfibs' records games played on FIBS and filters the textual output.
Therefore the setting boardstyle must be set to 3 and the stdout of
FIBS must be piped through 'rfibs'. The correct call for 'rfibs' is:
'telnet <...> | rfibs <args>'.
Two additional FIBS-commands are available with 'rfibs':
- bstyle <b>: switches the board display. Additional board displays can be
created and added in the source 'boards.c' and 'fibs.h'. <b> currently
ranges from 0 to 1.
- switch: switches the player and the board.
- recinit: stops recording of the actual game.
- comment <comment>: A comment is inserted into the record file.
Every started, resumed or watched game will be stored in
<record file>.
Every talk of the players or watchers will be included in the game as a
comment.
The argument '-c' suppresses recording the talks as comments.
The created file has the same syntax like the output of the FIBS-command
'oldmoves'.
Bugs:
The prompt '>' is missing
Prompts for login and password are missing, too.
sfibs { <gamefile> [-g] | -p } [-y<b>] [-s]
'sfibs' shows a recorded game given in <gamefile>. The gamefile can
be created with 'rfibs' or with the FIBS-command 'oldmoves'.
Argument '-g' suppresses interaction: The whole game is printed.
Argument '-y' gives the boardstyle. The available boardstyles are the same
as in 'rfibs'.
Argument '-s' puts player O on the downside. Default is player X on the
downside.
Argument '-p' doesn't show a game. It enables the user to create and to
print his own positions.
'rfibs' and 'sfibs' are given in the following files:
readme, makefile, fibs.h, sfibs.c, rfibs.c, boards.c
An "as is" DOS Port by Robin Davies (FIBS/W author) is available for
ftp from:
ftp://ftp.cybercom.net/pub/users/damish/backgammon/rfibs.zip
This version will work with output saved using the '>' command for saving
matches from FIBS/W. The file currently contains executables only.
LaTeX Style for BG Positions and
Games I have just finished the first version of a LaTeX style to print out
positions and matches.
The main features are:
- Boards produced with a special font made with
metafont, no inclusion of Postscript files needed (i.e. every dvi previewer
should be able to display the boards).
- Two different environments for single positions and
complete matches.
- Board layout customizable in both environments.
- Automatic generation of the current board at
arbitrary places in the game environment. Only the moves have to be entered,
the state of the board is maintained internally by the style.
I have uploaded my LaTeX package to ftp.dante.de as a
CTAN submission and was told that it's installed under
macros/latex209/contrib/backgammon
It should be available on every CTAN Server (these are
if I am right informed: ftp.dante.de, ftp.shsu.edu and ftp.tex.ac.uk)
Joerg
BOA/386 Bearoff analyzer From Harold
Wittmann
wittmann@fmi.uni-passau.de
I have written a piece of backgammon software that gives
you the winning probability for bearoff positions.
BOA/386. It's inexpensive shareware. Try it!
--------------------------------------------
Never again rollout bearoffs!
Here is what FILE_ID.DIZ says:
+-------------------------------------+
| BOA/386 v1.1 |
|-------------------------------------|
| A Backgammon Bearoff Analyzer: |
| - gives probability of each side |
| winning, cubeless (both side must |
| have all their men in the inner |
| board) |
| - shows proper way to play any roll |
| of dice |
| - very fast and accurate |
| - less than 1MB HD-space |
| - MS-DOS, 386SX+ required |
|-------------------------------------|
| BOA/386 is inexpensive shareware. |
| Only 20$ registration fee. |
+-------------------------------------+
Available for anonymous ftp from:
ftp://ftp.cybercom.net/pub/users/damish/backgammon/
Please note that the file is over 800k.
Section D: RESOURCES
Below is a list of backgammon clubs in North America.
It was taken from the January/Febrary 1995 issue of the Chicago Point
newsletter. It may be copied for noncommercial purposes as long as you give full
credit to "CHICAGO POINT, 3940 W. Bryn Mawr Ave., Suite 504 Chicago, IL
60659-3128"
Information for this listing has been obtained directly
from the featured backgammon clubs. Changes are inevitable and we suggest that
you always call before attending. As a service to our readers, CHICAGO POINT
would like to maintain an accurate listing of Backgammon Clubs In North America.
Please send information including updates, club closings, and new clubs to:
Contact: Chicago Point
Bill Davis, Editor
3940 W. Bryn Mawr Ave., Suite 504
Chicago, IL 60659-3128
Phone: 312 583-6464
Fax: 312 583-3264
(FORMAT)
CLUB NAME....................... LOCATION..................
CITY & STATE........ MEETING ON................ TIME...
CONTACT............. TELEPHONE...
MAIL INQUIRIES TO...................................
(WEST)
Backgammon Club of San Diago Java House
Delmar, CA Sunday 2:00 PM
Mike Fujita 619/294-2007
P.O Box 178119; San Diago, CA 92177
Backgammon Club of San Diago II Forno
La Jolia, CA Wednesday Monthly 6:30 PM
Mike Fujita 619/294-2007
P.O Box 178119; San Diago, CA 92177
San Francisco Backgammon Golden Gate Grill
San Francisco, CA Tuesday 7:45 PM
Doug Adsit 415/931-4600
3200 Fillmore Street; San Francisco, CA 94123
San Francisco Bridge & BG Club Clubroom
San Francisco, CA Wed., Fri. 9:00 PM
Augie Hunt 415/776-6949
777 Jones Street; San Francisco, CA 94109
Gammon Associates Grand Slam Bridge & BG Club
Woodland Hills, CA Tuesday 7:00 PM
Patrick Gibson 818/901-0464
7641 Orion Avenue; Van Nuys, CA 91406
Gammon Associates Grand Slam Bridge & BG Club
Woodland Hills, CA Sunday 7:00 PM
Patrick Gibson 818/901-0464
7641 Orion Avenue; Van Nuys, CA 91406
Prime BG Club of Las Vegas Jockey Club
Las Vegas, NV Tuesday 7:00 PM
Howard Markowitz 702/893-6025
2620 S. Maryland Pkwy; Box 230; Las Vegas, NV 89109
No. Nevada Backgammon Assn. Rapscallion Seafood House
Reno, NV Thursday 7:30 PM
Jim Allen 702/329-1227
449 W. Plumb Lane; Reno, NV 89509
Oergon Backgammon Club Lacey's
Lake Oswego, OR 1st & 3rd Tuesday 7:00 PM
Henry Moss 503/636-6258
2360 Greebtree Road; Lake Oswego, OR 97034
Pacific NW Backgammon Assn. Shakey's Pizza
Bellevue, WA Monday 7:00 PM
Chuck Breckenridge 206/778-8181
18204 Olympic View Drive; Edmonds, WA 98020
Puget Sound Backgammon Assn. European Connecktion
Seattle, WA Tuesday 7:30 PM
Guy Thurber 206/244-6737
428 SW 127th Street; Seattle, WA 98146
(CENTRAL)
Bloomington-Normal BG Club Ride The Nine
Bloomington, IL 1st/3rd/5th Tues. 6:15 PM
Lane O'Connor 309/454-1947
108 Riss Drive; Normal, IL 61761
Games People Play Alexander's
Chicago, IL Monday 6:30 PM
J.A. Miller 312/768-5523
P.O. Box 8630; Chicago, IL 60680
Chicago Bar Point Club Golden Flame
Chicago, IL Tuesday 6:15 PM
Bill Davis 312/338-6380
2726 W. Lunt Avenue; Chicago, IL 60645
Chicago Bar Point Club Braxton Seafood Grill
Oak Brook, IL Sunday bimonthly 12:30 PM
Peter Kalba 312/252-7755
2510 W. Iowa Street; Chicago, IL 60622
Pub Club Crickets Pub & Grill
Glendale Heights, IL Monday 7:30 PM
V.W. Zimnicki 708/924-8632
P.O. Box 72216; Roselle, IL 60172
Prime BG Club of Chicago TJ's Lounge/Radisson Hotel
Lincolnwood, IL Friday 7:00 PM
Joann Feinstein 708/674-0120
8149 Kenton; Skokie, IL 60076
Central Illinois Backgammon Club Cummins Family Restaraunt
Peoria, IL Thursday 6:30 PM
Ed Bauder 309/686-6662
1115 E. McClure Avenue; Peoria, IL 61603
Sangamon Valley BG Assn. Parker's Sports Bar
Springfield, IL 2nd & 4th Tuesday 6:00 PM
Sangamon Valley BG Assn. Parker's Sports Bar
Springfield, IL 1st Sat./Nov.-Apr. 11:30 AM
Randy Armstrong 217/528-0117
2012 N. 20th Street; Springfield, IL 62702
Winnetka Backgammon Club Winnetka Community House
Winnetka, IL Wednesday 7:00 PM
Trudie Stern 708/446-0537
4200 W. Lake 302C; Glenview, IL 60025
Summit City Backgammon League Alumni Club
Ft. Wayne, IN Wednesday 7:00 PM
Ken Bruck 219/639-6898
P.O. Box 6546; Fort Wayne, IN 46896
Hoosier Backgammon Club Spats
Indianapolis, IN Thursday 7:00 PM
Butch Meese 317/845-8435
7620 Kilmer Lane; Indianapolis, IN 46256
Flint Area Backgammon Club Ramada Inn
Flint, Ml Thursday 7:00 PM
Carol Joy Cole 810/232-9731
3003 Ridgecliffe Drive; Flint, MI 48532
Plymouth Backgammon Club Box Bar & Grill
Plymouth, MI Wednesday 7:30 PM
Dean Adamian 313/981-5706
42954 Barchester; Canton, MI 48187
Cavendish North BG Club Clubhouse
Southfield, MI Daily except Sun. 1:00 PM
Joe Sylvester 810/642-9616
30065 Greenfield Road; Southfield, MI 48076
Minneapolis Backgammon Minneapolis Athletic Club
Minneapolis, MN Monday 5:00 PM
Fred Kalantari 612/682-1716
4701 Valley View Road; Edina, MN 55424
Cleveland Area Backgammon Boulevard Sports Tavem
Cuyahoga Falls, OH Wednesday 6:30 PM
Irv Taylor 216/663-7332
P.O. Box 28515; Cleveland, OH 44128
Dallas Backgammon League Scoreboard
Addison, TX Wednesday 7:45 PM
Rich Weaver 214/620-7462
2682 Hearthstone; Dallas, TX 75234
Austin Backgammon Assn. Bombay Bicycle Club
Austin, TX Monday 7:30 PM
Jackie Seiders-Smart 512/261-8518
30 Hightrail Way, Austin TX 78738
American Backgammon Club Vickery Feed Store
Dallas, TX Sunday 6:00 PM
Kati Pratt 214/827-8403
5631 Ellsworth; Dallas, TX 75206
Houston Backgammon Club Abdallah's
Houston, TX Tuesday 7:00 PM
Jack Butler 713/774-9439
5931 Reamer Street; Houston, TX 77074
San Antonio Backgammon Dad's
San Antonio, TX Friday 8:00 PM
Marcel Mommers 210/606-0025
3812 Greenridge Drive; Cilbo, TX 78108
Milwaukee Backgammon Club Gas Lite North
Milwaukee, WI Wednesday 7:00 PM
Marv Amol 414/355-8805
9031 N. 70th Street; Milwaukee, WI 53223
Milwaukee Backgammon John Hawks Pub
Milwaukee, WI Sunday Bimonthly 11:30 AM
Merril Schrager 414/463-2498
9043 W. Grbaosa Druve; Milwaukee, WI 53225
(NORTHEAST)
Beltway Backgammon Club Promenade Cardroom
Bethesda, MD 2nd & 4th Sunday 12:00 N
Barry Steinberg 301/530-0604
5712 Quebec Street; Benwyn Hts. MD 20740
bkgm@aol.com
Cavendish Club of Boston Clubroom
Brookline, MA Thur. & Sat. 1:00 PM
Carl Saldinger 617/734-2230
111 Cypress Street; Brookline, MA 02146
New England Backgammon Club Sheraton Commander Hotel
Cambridge, MA Sunday monthly [Sep-Jun] 12:30 PM
Andy Latto 617/374-2537 (days) 617/784/6114 (eves)
156 Massapoag Avenue; Sharon, MA 02067
andyl@harlequin.com
New England Backgammon Club Sheraton Commander Hotel
Cambridge, MA Monday 7:00 PM
New Jersey Backgammon Assn. Best Western Oritani Hotel
Hackensack,NJ Tues. & Fri. 7:45 PM
Ron Whitney 201/833-2915
279 Glen Court; Teaneck, NJ 07666
Ace Point Backgammon Club Clubroom
New York, NY Daily 3:00 PM
Michael Valentine 212/753-0842
41 E. 60th Street; New York, NY 10022
Coterie Clubroom
New York, NY Daily 1:00 PM
Louise Goldsmith 212/371-5151
Private club. Telephone for information.
New York Chess & Backgammon Office Building
New York, NY Daily, Tour. Sunday 12:00 N
Steve Manning 212/302-5874
120 W. 41st Street 3; New York, NY 10036
Saratoga Backgammon Club Waterfront Restaraunt
Saratoga Springs, NY Friday 1:00 PM
Lee Hoge 518/584-1714
P.O. Box 563; Saratoga Springs, NY 12866
Greater New York BG Club Woodbury Ramada Inn
Woodbury, NY Sunday monthly 12:30 PM
Dr. Bob Hill 718/341-3779
194-22 115th Road; Jamaica, NY 11412
Cavendish Club of Philadelphia Clubroom
Philadelphia, PA Mon, Wed, Thu, Sat 1:00 PM
Ken Relver 215/878-5777
3801 Conshohocken Avenue; Philadelphia, PA 19131
Pittsburgh Backgammon Assn. Murphy's Tap Room
Pittsburgh, PA Tuesday 8:30 PM
Steve Hast 412/823-7500
3560 Ridgewood Road; Pittsburgh, PA 15235
(SOUTH)
Suncoast Backgammon Assn. New York, New York Lounge
Clearwater, FL Monday 7:00 PM
Drew Giovanis 813/726-1398
25350 U.S. Hwy 19 N. 67; Clearwater, FL 34623
South Florida Backgammon Big Apple Sports Club
Ft. Lauderdale, FL Sunday 1:00 PM
Elayne Feinstein 305/785-1282
2621 NE 7th Terrace; Pompano Beach, FL 33064
Backgammon Club of N.W. Florida Olde English Pub
Lynn Haven, FL Tuesday 7:30 PM
Rick Bieniak 904/773-2013
P.O. Box 416; Wausau, FL 32463
Orando Backgammon Coach's Locker Room
Orando, FL Tuesday 7:30 PM
David Thomas 904/736-2844
P.O. Box 803, Deland, FL 32721
Backgammon Society of Sarasota Bath & Raquet Club
Sarasota, FL Tuesday 7:00 PM
Frank Shank 813/792-3992
5320 86th St. W.; Brandenton, FL 34210
Atlanta Backgammon Association Cabo Wabo & Grill
Atlanta, GA Wednesday 7:00 PM
Dave Cardwell 404/682-1969
P.O. Box 956547; Duluth, GA 30136
Atlanta Backgammon Association CafFiends Coffee Bar
Atlanta, GA Sunday Monthly 7:00 PM
Dave Cardwell 404/682-1969
P.O. Box 956547; Duluth, GA 30136
Louisville Backgammon Club Encore Restaurant
Louisville, KY 2nd & 4th Tuesday 6:30 PM
Quint McTyeire 502/896-9783
4906 Crofton Road; Louisville, KY 40207
Raleigh Backgammon Club Western Bowling Alley
Raleigh, NC 1st & 3rd Friday 7:30 PM
Frank Bommarito 919/552-2291
401 St. John Court; Holly Springs, NC 27540
(CANADA)
Vancouver Backgammon Jo-Anne's Place
Vancouver, BC Tuesday 8:00 PM
Marty Jensen 604/688-8317
1339 Burnaby St. 402; Vancouver, BC V6E 1R2; CANADA
Nat'l Capital Backgammon Club Lunergan's Pub
Vanier, ON lst Sun./Sep.-Jun. 12:30 PM
Eden Windish 613/741-2530
396 Talbot Street; Ottawa, ON KlK 2N6; CANADA
Le Gammon Clubroom
Montreal, QB Daily 3:30 PM
Michel Medifti 514/845-8370
552 St. Catherine East, Montreal, OB H2L 2E1; CANADA
From: mau@world.std.com (Michael A Urban)
Date: 19 Oct 1993
Frequently, membership fees are waived for initial participants.
For complete details, contact the club of interest.
Cavendish Club 617-734-2230
111 Cypress St.
Brookline, MA 02146 USA Fee: $150/year
The Cavendish runs chouettes on Thursday evenings and Saturday
afternoons. The club also has duplicate and rubber bridge.
New England Backgammon Club
c/o Sheraton Commander Hotel
16 Garden Street
Cambridge, MA 02138-3609 USA Fee: $35/year
President:
Andy Latto 617/374-2537 (days) 617/784/6114 (eves)
156 Massapoag Avenue; Sharon, MA 02067
andyl@harlequin.com
The NEBC runs weekly Monday tournaments starting at 7pm and monthly Sunday
tournaments beginning at 1pm. No smoking is permitted in the tournament room.
The NEBC publishes,
"Anchors", a monthly newsletter.
A homepage for the NEBC is under construction:
http://www.cybercom.net/~damish/backgammon/nebc/nebc.html
[Note: There are no Sunday tournaments in July or
August]
[Note: Some Sunday tournaments start at noon.
Nov and May in 94/95 season] [Note: There are no Monday tournaments During the
summer of 1995]
The Danish BG Federation Phone: (+45 39 40 06 07)
The Danish Backgammon Federation is maintaining two
extensive national ratinglists (all matches and tournament matches). As far as I
recall we are currently rating more than 50,000 matches a year !!
Furthermore the DBgF is offering a variety of things
stretching from a monthly magazine, live tournaments EVERY DAY, international
tournaments (NORDIC OPEN every Easter) and many more things.
For your information: The DBgF is a democratically
governed organisation. We encourage all other nations to form national
organisations. It is the only way if we really want to promote the game in a
serious manner.
Erik M. Gravgaard
President of
the Danish BG Federation
erikg on FIBS
[95-03]
In your FAQ you have
details of backgammon clubs ... can you include ours please? Several fibsters
come to it regularly.
Rose and Thistle, Argyle Road, Reading, UK. Weekly
Wednesdays from 8:30 pm. Informal - no officers, but contact number: Marina
01734 612814
Thanks. :-)
--
* Marina Smith * Reading, U.K. *
marina on FIBS.
From: Jake Jacobs
Newsgroups:
rec.games.backgammon,rec.gambling
Subject: New
Backgammon Club in Chicago
Date: 4 Dec 1995 16:48:00
GMT
In January, 1996, Dean Muench will be opening the
Chicago Board Of Backgammon. The CBOB will be a private club for backgammon and
other games. Address is 175 w. Jackson (right near the CBOT). It will be open
Monday through Friday from 12 noon, and on weekends for special events. I can
forward inquiries to Dean till he gets a phone #. I'm at
itaewon@interaccess.com.
The CBOB will hold a grand opening tournament on 1/6/96
at 12 noon. The address is 175 w. Jackson, suite 1103. The phone number from
that date onward will be 312-922-0404.
From: pfmcm@worm.hooked.net (Paul McMillan)
Newsgroups: rec.games.backgammon
Subject: New Weekly BG Site in San Francisco
Date: 27 Feb 1995 15:55:35 -0800
The Baja Cantina will begin hosting weekly BG tournaments on Wednesdays at
8:00 p.m. starting March 8, 1995.
The format will largely resemble that of the tournaments
formerly held on Tuesdays at the Golden Gate Grill.
The Baja is located across the street from the Golden
Gate Grill at 3154 Fillmore, corner of Greenwich. The phone number there is 415
885-2252.
Please bring your boards.
Hi Mark! On my recent trip to Sydney, I visited the bg
club that has been running there 12 years. They said to put it on the FAQ, so
here are the details:
The Clocktower pub
Corner of
Crown St. and Nixon St.
Tuesday nights.
* Marina Smith * Reading, U.K. *
From: "Dr. Linton Hutchinson" (webuniv@iag.net)
Newsgroups: rec.games.backgammon
Subject: Re: North American clubs on WWW
Date: 28 May 1995 03:35:26 GMT
The Orlando Backgammon meets on Tuesday 9:30 to whenever. The location
is on 436 just past I-4. First shopping center on the right. Behind TGI
Friday's at the Coaches Lockerroom. Upstairs.
The Tyson's Corner Backgammon Club meets every Monday at 7:00 PM at
Mr. Smith's Restaurant, 8369 Leesburg Pike (Route 7) in the Pike 7
Shopping Center. It's in northern Virginia near Washington, DC.
call Bill Pow at (703) 549-1808 or (301)227-1915
[Submitted by Mel Leifer]
Backgammon clubs around the world may be obtained from:
http://www.statslab.cam.ac.uk/~sret1/backgammon/clubs/index.html
- Look for ads in backgammon newsletters.
- Check backgammon clubs.
- Sometimes announcements are posted to the news group
rec.games.backgammon.
- Various WWW pages are now posting schedules.
Newsletter of the New England Backgammon
Club
Monthly except July, usually 8 pages, two devoted
to local news with remaining to analytical material and backgammon related
stories with minimal advertising.
Subscription: USA/Canada/Mexico: $15/year.
Overseas: $25/year (check drawn on U.S. bank).
Contact: NEBC
c/o Sheraton-Commander Hotel
16 Garden Street
Cambridge, MA 02138-3609
Lately this has been published
quarterly in a combined German/English edition.
Vertrieb (sales/marketing)
Edith Johanni
Emil-Nolde-Str. 26
90455 Nurnberg
Tel.: 0911/883253
Redaktion (Editor)
Harold Johanni
Hochstr. 7
90429 Nurnberg
Tel.:0911/269567
Bi-monthly, 24 A5 sized pages. in Danish. It
is a newsletter/magazine for a BG club in Copenhagen. 4-6 pages are directed
towrds the members of the club, the rest is general analysis, commentated
matches, articles, problems etc. There is a quiz a la Inside BG's quiz. 4
problems each issue, with the answers taking 4-6 pages.
Subscription: Scandinavia Danish kr. 120
Europe Dkr. 140
Overseas (USA/Canada) Dkr. 160
Contact: Asger Kring (proj13@srv.imsor.dth.dk)
Lykkesholms Alle 4B,3 th.
1902 Frederiksberg C.
Denmark
tlf: 3131 1439
A Prime Source of Backgammon Information
Monthly newsletter, 10 pages:
Subscription: USA/Canada/Mexico: $25/year.
Overseas: $35/year airmail in USD check drawn on
U.S. bank.
Contact: Chicago Point
Bill Davis, Editor
3940 W. Bryn Mawr Ave., Suite 504
Chicago, IL 60659-3128
Phone: 312 583-6464
Fax: 312 583-3264
Contact: European Backgammon News
Apartado 81
E-04630
Garrucha (Almeria)
Spain
FAX: 34/68-438347
Subscription: $30/12 issues in Europe. $40 elsewhere.
Published since July 1993 by Martin de Bruin.
Monthly newsletter, 10 pages: Problem analysis, book and
software reviews, tournament schedules and complete results, local, national and
international backgammon news and views. Full page catalog of backgammon
merchandise.
Subscription: USA/Canada/Mexico: $20/year or $200 lifetime.
Overseas: $25/year or $250 lifetime subscription.
Contact: Carol Joy Cole, Editor
3003 Ridgecliffe Drive
Flint, Michigan 48532-3730 USA
Phone/Fax: 810-232-9731.
Members magazine for the Danish Backgammon
Federation.
A5 sized. Published 6 - 8 times a year with
a circulation between 1,800 and 10,000 (10,000 during Carlsberg Backgammon Cup).
32 to 44 pages with articles, rating lists, tournament
invitations and reviews, general BG theori, "ask the experts"-coloumn, annotated
matches etc. Mainly in Danish, with occasional English articles.
Subscription can also be obtained by non-members for DKK
180 per year (US$ 30).
Advertisement: Please contact
Chris Ternel for prices.
Editor:
Chris Ternel
Danish Backgammon Federation
Gersonsvej 25
DK-2900 Hellerup
DENMARK
Tel. +45 39 40 06 07
Fax. +45 39 40 01 44
or you can contact:
Erik Gravgaard, president of DBgF
e-mail: erikg@inet.uni-c.dk
Bi-monthly, 8 pages: Articles/problems plus
very issue non-annotated matches of todays best players.
Subscription: USA: $12/years Canada/Mexico: $14/year.
Overseas: $16/year (cash or check drawn on US bank.)
Contact: Butch & Mary Ann Meese
Hoosier Backgammon Club
7620 Kilmer Lane
Indianapolis, IN 46256 USA
Tel: 317.845.8435
E-mail: hbc@ix.netcom.com
Bi-monthly, 24 pages: Technical magazine with quizzes,
articles and annotated matches by the best backgammon players today.
Subscription: USA: $40/years
Canada/Mexico and oversea ground: $45/year.
Overseas airmail: $60/year (US funds).
Contact: INSIDE BACKGAMMON
P. O. Box 294
Arlington, MA 02174 USA
(617) 641-2091
Norpunkt is the magazine of the Norwegian
Backgammon Federation (Norges Backgammon Forbund). It's published quarterly, but
only written in Norwegian.
A sample of articles written in Norwegian appears at WWW
site: http://www.oslonett.no/home/swelle/norpunkt.html
The WWW page for the Norwegian Backgammon Federation is:
http://www.oslonett.no/home/swelle/NBgF.html
For Further information send E-mail to Stein Welle at swelle@oslonett.no
Are there any other backgammon newsletters or magazines out there? Please help
to keep the faq up to date. (thanks!)
(From rec.games.backgammon)
From: hunter@work.nlm.nih.gov (Larry Hunter)
Subject: Bibliography
Date: 21 May 92 20:11:35 GMT
Sender: usenet@nlm.nih.gov (usenet news poster)
One more for the FAQ. Before Marty Storer left the list, I pestered him
for an annotated bibliography. He came through in grand style. Here it is:
Must have:
- Backgammon
- Paul Magriel, NY Times/Quadrangle Press, New York
1976.
- The best introduction to the game. Covers basic
checker play very well. If you read and thoroughly understand this book,
you'll play a decent game. Weaknesses--skimpy treatment of the doubling cube.
- Genud vs Dwek: The 1981
World Backgammon Championship_ (or similar title)
- Bill Robertie, The GAMMON PRESS,
Arlington, Mass. 1982.
- Very thorough coverage of the 25-point finals of the
1981 Monte Carlo tournament. Goes into quite a bit of detail about ins and
outs of match play. Excellent section on backgames. I've referred to this as
Robertie(red) since it has a red cover 8-).
- Backgammon With The
Champions
- Kent Goulding, ~1980-82.
- Series of annotated matches between good players.
Forget how many in all. Excellent material, giving very good insight into how
top players think. Commentary by Goulding, often in collaboration with Kit
Woolsey; both of these guys are very, very strong players. Let's see, the
matches are Seidel vs. Hodis; Magriel vs. Sconyers; Genud vs. Posner; Pasko
vs. Motakhasses; two (?) 5-point matches in one volume: Lester vs. Horan and
Woolsey vs. Pasko; Robertie vs. Senkiewicz; Goulding vs. Maxakuli; Dwek vs.
Chafetz; Ballard vs. Lubetkin; Eisenberg vs. Magriel(?); and more I can't
remember. I can't recommend this series too highly (though Genud vs. Posner
was a lousy match).
- Advanced Backgammon (2nd
edition; two volumes)
- Bill Robertie,
- The GAMMON PRESS, Arlington, Mass. '91. I
haven't seen this yet--only the first edition of one volume. Series of
problems, giving very good introduction to truly advanced concepts. Errors in
first edition are supposedly corrected. The first edition is what I call
Robertie(blue); the second is Robertie(white).
- Backgammon Times ,
all back editions.
- This was a very good backgammon newspaper that was
around in about '82-'83. A lot of interesting articles by top players and
analysts. Probably hard to get these days.
- Reno 1986
- Bill Robertie, The Gammon Press, Arlington, Mass.
1987.
- Two annotated matches from the very strong Reno
tournament of '86. Semifinal match is between Nack Ballard and Mike
Senkiewicz; an excellent match, well annotated. Finals between Ballard and
Howard Markowitz. The book is in quiz format, so you can test your skill
against Ballard's (well, kind of: Ballard had to find his moves over-the-board
under great pressure--nothing like the finals of a big tournament to get the
adrenalin flowing!). I've only found a couple of mistakes in the annotations.
This book is referred to as Robertie(yellow).
- World Class Backgammon,
Move By Move
- Roy Friedman, 1989 or 90;
- forget other publication info. Annotated matches
between Robertie and ``international star'' Rick Barabino (Barabino is strong,
but ``international star''--I dunno...). Three 9-point matches with some
excellent games (check out the second game of the first match particularly).
Annotations are very good; Friedman put a lot of work into rolling out many of
the diagrammed positions. The annotation style is terse; Friedman takes a very
scientific approach.
- Vision Laughs at Counting
(two volumes)
- Danny Kleinman, ~1978.
- -all other material by Kleinman is "must have"--write
to him at 5312-1/2 Village Green, Los Angeles, CA 90016 and tell him I sent
ya.
- Seminal work on match play, money play, doubling
cube, races, and more. Kleinman is very prolific. His analyses are often more
mathematical than the average reader can handle, but Real Mathematicians [tm]
and even the layperson with math aptitude shouldn't be fazed. A Real
Mathematician wouldn't call Kleinman's math "deep", but it sure is accurate,
and you won't find anything similar anywhere else. He does the important work
of formulating the right problems the right way, where many others couldn't.
Drawbacks: his books are self-published with lousy
layout and graphics. He's supposedly not that great a player (I've never seen
him play), so his analyses often lack the world-class insight into the thought
processes of the strong practical player that you could get from a Goulding or
a Robertie. In particular, his middle-game intuition seems less than
world-class. But these drawbacks are more than made up for by the wealth of
information in his books, which I still haven't completely soaked up after
many years. Kleinman is a subtle thinker and a meticulous analyst of the
countable, and he does a lot to develop backgammon "vision." His stuff is
often uproariously funny, but sometimes one gets impatient trying to filter
out what's relevant to the practical player from the humor.
I repeat--all his books are "must have's" for the
serious player. They're a bit expensive since I think he bears all the
production costs himself, but for the serious player they're worth every cent.
Pretty Good Books But Not "Must Have's":
- Backgammon For Profit
- Joe Dwek, Stein and Day, New York 1975 (out of print)
- Problems that would now be considered fairly basic.
Almost all solutions are right. Tables of replies to opening moves show how
badly people played in 1975.
- Paradoxes and
Probabilities
- Barclay Cooke, Random House, New York 1978.
- This is almost a "must have." 168 problems, most of
which are very interesting. Current thinking is that solutions to about a
third of them are wrong, but the analysis gives very good insight into how
Cooke, a first-generation world class player, thought about backgammon.
- The Doubling Cube In
Backgammon
- Jeff Ward, Aquarian Enterprises, San Diego 1982.
- Goes into basic doubling-cube concepts and gives some
benchmark positions with equities derived from rollouts. Gives some bearoff
tables, etc. Analysis of benchmark positions is pretty good but sometimes
skimpy; Ward only admits to having done 100-200 rollouts to derive his
equities. Worth having.
- Backgammon Master Games
- Bill Kennedy and Chuck Papazian, 1982 (forget other
publication info).
- Annotated games and positions from master match play.
Analysis is largely based on intuitive concepts, and isn't well grounded in
match-equity considerations etc. Not well supported by rollouts; a fair amount
of errors, but the analysis overall is pretty sound.
Other books that I've read aren't worth much, including
Competitive Backgammon Vol. II, Mike Labins,
Marty Storer, and Bill Tallmadge, Competitive Backgammon Publications, Syracuse
1981. (It was good for the time but would be considered lousy now.)
As I mentioned before, you can reach Gammon Press at
(617)641-2091, fax: (617)641-2660 or PO Box 294 Arlington, MA 02174 USA
Larry
--
Lawrence Hunter, PhD.
National
Library of Medicine
Bldg. 38A, MS-54
Bethesda. MD 20894
(301) 496-9300
(301) 496-0673 (fax)
hunter@nlm.nih.gov (internet)
[Note: Fax number edited to reflect the current number.]
Article: 1666 of rec.games.backgammon
Newsgroups: rec.games.backgammon
From: johnsson@sara.cc.utu.fi (MIKA JOHNSSON)
Subject: BG BOOKS INFO
Date: Thu, 1 Jul 1993 10:20:19 GMT
HI !
Many people (last Snoopy) have asked about good BG
books, well here is and answer that i got from John Bazigos (Doc), when I asked
him about books
-Mika
BG BOOKS BY JOHN BAZIGOS
The two best introductory books are Paul Magriel's ``Backgammon'' (New York Times Quadrangle Press; New
York, NY; USA; 1976) and Enno Heyken's and Martin B. Fischer's ``The Backgammon Handbook'' (The Crowood Press;
Ramsbury, Marlborough, Wiltshire SN8 2HE; Great Britain; 1990).
The advantages of Magriel's ``Backgammon'' are, first, Magriel was a
clear-minded, distinguished mathematician at the top of the backgammon world
when he wrote it; second, it was the only truly analytic book on backgammon
since Oswald Jacoby's and John R. Crawford's ``The
Backgammon Book''; third, it rendered all backgammon texts preceding it
(including ``The Backgammon Book''), and even
some subsequent backgammon texts, obsolete as introductory texts; fourth, it
systematically elucidates backgammon strategy, from fundamental to intermediate
to advanced; fifth, it does great justice to its topics in its well-diagrammed
over-400 pages; and sixth, it has passed the test of time as an introductory
text, having been commonly referred to as ``The Bible'' of backgammon. Its
disadvantages are, first, some important details of some advanced topics (e.g.,
desirable back-game points), and even some major points of some
beginning/intermediate topics (e.g., tradeoffs between positional and racing
equity) are obsolete; second, the prose, though very readable, is structurally
and stylistically weak; third, the text has been out-of-print since some time
last year, though is well worth a search of *all* your local used/out-of-print
bookstores; and fourth, though the publication price was $24.95, the only
mail-order list on which I have found it prices it at $80, which makes a
used/out-of-print bookstore an even better source -- since it is typically
discounted to about $15 there, in my experience (here in the San Francisco Bay
Area).
The advantages of ``The
Backgammon Handbook'' are, first, like ``Backgammon'', it systematically elucidates
backgammon strategy; second, it contains the complete score, with some
annotations, of the very illuminating, 26-game match between two-time World
Backgammon Champion and leading bg theorist Bill Robertie and now-inactive [as
of this writing] international master Nack Ballard (Reno, 1987) that the former
described as ``...perhaps the most interesting one I've ever played in my
life!''; and third, it is still in print with a publication price of about $35.
Its disadvantages are, first, Heyken --though an International Master in chess--
does not have an international backgammon rating, and Fischer does not have a
master rating in backgammon; second, it contains only about 60% as much text as
``Backgammon'', while not being significantly
terser; and third, the authors' lack of qualifications is evidenced in some of
their misleading and/or naive analyses.
I think that you should search your local, or even not
quite local, used/out-of-print bookstores for ``Backgammon'', and pay up to about $50 for it --
though if you find it in such a store, it is likely to be discounted to about
$15; and then, if you cannot find it at a reasonable price, buy and read ``The Backgammon Handbook'' -- after which your time
won't be best spent reading Magriel soon thereafter.
[ Note:
The Backgammon Handbook is still in print and can be ordered direct from
the (very small) publisher:
The Crowood Press
Ramsbury
Marlborough
SN8 2HE, U.K.
Or from any decent bookshop. The price is UKP 15.95 ]
> are you familiar with Danny Kleinmans books,
I have read most of his ``magnum opus'' ``Vision Laughs at Counting'', which contains much
sound advice on the practical aspects of bg play (e.g., sections on bg hustlers,
bg cheaters, chouette money management), seminal advice on handling the doubler,
and even a few unprecedented mathematical characterizations of certain aspects
of certain positions (e.g., how many pips to penalize a player for having one or
more checkers on the bar).
> are they good ?
``Vision Laughs at
Counting'' is generally insightful and often very amusingly written, but
not suitable as an introductory text, sometimes obsolete, and sometimes simply
wrong; and though it is the only text by Kleinman that I have read, I have good
reason to believe that that judgment applies to Kleinman's other texts, as well.
Ok; then after finishing ``Backgammon'' or ``The
Backgammon Handbook'', study Jeff Ward's ``The
Doubling Cube in Backgammon'' -- which has long been offered through Carol Joy Cole.
Magriel's ``Backgammon'' routinely used to be, and sometimes
still is, referred to as ``the Bible (of backgammon)''; but since the
publication of Robertie's three books on backgammon --i.e., ``Lee Genud vs. Joe Dwek'' (1982), ``Advanced Backgammon'' (1984 and 1991, the latter
edition in two volumes), and ``Reno, 1986''
(1987)-- I think that it's more appropriate to refer to ``Backgammon'' and collectively those three as the
Old and New Testaments of backgammon, respectively. Given that you have already
finished studying ``The Backgammon Handbook''
and ``The Doubling Cube in Backgammon'', I
think that you should read one or more books of Robertie's ``New Testament''
fairly soon after finishing Roy Friedman's ``World
Class Backgammon, Move-By-Move'' -- which I, also, recently received a
copy of from Carol Joy
Cole, and is the backgammon book that I intend to read next.
Well, from the quality perspective, I was significantly
more impressed with it when perhaps the only bg literature I had read was
typical junk from the 1970s (i.e., Bruce Becker's monumentally horrible ``Backgammon for Blood'', and Barclay Cooke's
often-misleading ``The Cruelest Game'' and
slightly-better ``Championship Backgammon''),
``The Backgammon Book'', and Magriel's ``Backgammon''; and from the price perspective, the
decision is strictly yours, though I hereby make the following three
interrelated claims:
- If you read enough backgammon books, there will quite
possibly come a time when ``Vision Laughs at
Counting'' will be the best book for to read next to improve your
technique maximally.
- You are probably at least seven books from that
point: ``World Class Backgammon,
Move-By-Move'', the four volumes of backgammon's New Testament, and
both volumes of Kent Goulding's ``Backgammon With
The Champions'' are presently better for that purpose (and you can
perhaps most profitably read them in that order).
- ``Vision Laughs at Counting'' is the most
entertaining instructional backgammon book that has been published to date.
> BTW are there other good bg newspapers or magazines
?
Last year was an unprecedentedly good one for backgammon
periodicals, in that it saw the first issues of what I strongly believe were and
still are the two best periodicals for backgammon theory ever --i.e., Bill
Robertie' and Kent Goulding's bi-monthly ``Inside
Backgammon'', and Roy Friedman's almost bi-monthly ``Leading Edge Backgammon''. The former is still
being published (I recently received my copy of the fourth issue of its second
volume), and publication of the latter was suspended at the end of last year
(due to some personal problems that Roy was having); but it was possible to
order either or both of them from Carol the last time I checked (Please inform
me if you need ordering information on either or both of them).
Those are the only three backgammon periodicals to which
I (have ever) subscribe(d), though that may change soon; more on that in a
forthcoming e-mail message from me.
> Do you know any technical papers about BG,
One of the best features of both ``Inside Backgammon'' and ``Leading Edge Backgammon'' is they consist mostly
of (what I would consider) technical papers on backgammon.
>I have read Keelers and Spencers "optimal doubling
in BG"
So have I, but I have also read a paper co-authored by
Zadeh, titled ``On Optimal Doubling in
Backgammon'', that explicitly rendered that paper obsolete. I'll provide
you with more information on both that and other technical papers from the 1970s
in a forthcoming e-mail message.
> and in one AI-magazine was an article about
Tesauros TD-gammon (about 20 p)
The second volume of ``Inside
Backgammon'' contains about one article per issue on TD-Gammon, two of
which document (recent) sessions that Robertie, Magriel, and at least one other
bg master had against it; more on that, also, in an forthcoming e-mail message
from me.
``How to play tournament
BG'' by Kit Woolsey
This is an excellent introduction to how play and
particularly cube handling varies in games. It shows how to compute push and
cash points, recube equity, how to figure gammon costs, etc. It gives Kit's
latest match equity chart and gives a method for remembering most of it fairly
well. If you play matches games and don't immediately recognize any of these
terms, I strongly suggest reading it.
-michael j zehr
Here is an outline of the book:
0. Introduction
1. Crawford Game Strategies
2. Post-Crawford Play
2.1. The Free Drop
2.2. Mandatory Doubling
3. The Two-Away versus Two-Away Score
4. The Match-Equity Table
5. Learning the Table
5.1. The Janowski Formula
6. Using the Table
6.1. Gain-Loss Tables
6.2. The Doubling Window
7. Initial Cube-Decisions at Various Scores
8. Redoubles and Cube-Leverage
9. Cube-Leverage in Gammonish Positions
10. Gammon Potential and Checker Play
11. Five Practical Examples
12. An Illustrative Game: Woolsey-Robertie, Reno 1993
-- John Bazigos (``doc'' on FIBS)
Backgammon by
Robin Clay $7.95 NTC Publishing Group
I was surprised to see this book for sale recently at a
local book shop. This book was 'skimmed' by two intermediate players, and both
immediately found that the some of the concepts and advice given were grossly
incorrect. One of these 'reviewers', went as far as to say: "If your opponent
says that he has just read this book, immediately raise the stakes!".
...Mark
In The Game Until
The End: Winning In Ace-Point Endgames by Bob Watchel
You've played an ace-point game; Your opponent is down
to his last few checkers. Should you run? Should you stay? If your opponent
wants to settle, what's the game worth? How aggressively should you try to pick
up a second checker?
If you don't know the answers to these critical
questions, you need this book. In Chapter 4 alone you'll discover the secrets of
the famous "Tino Road Position," an endgame so complicated that - once you know
how to play it - you can take the position from either side and win. Olympiad
Champion Bob Watchel has thoroughly analyzed hundreds of ace-point game
positions to generate a complete picture of what's really going on in these
common yet widely-misplayed situations.
Soft bound, 112 Pages. Level: Advanced and Serious
Intermediate. Available from The GAMMON PRESS. US$25 + Shipping
[From a flyer from The GAMMON PRESS]
Learning from the
Machine: Robertie vs. TD-GAMMON by Bill Robertie
For years, computer backgammon was a languishing
sideshow, with the best computer programs barely able to rise to the
intermediate level.
This all changed in 1991 with the emergence of
TD-Gammon, an experimental neural network program developed at IBM's research
labs. TD-Gammon taught itself to play, starting with a knowledge of the rules of
the game. After playing thousands of games against itself, the program reached
strong Open player level; within months, it became world-class. TD-Gammon plays
like a strong human player in many parts of the game. In some areas, it plays
quite unlike what has generally been accepted as "correct strategy" leading
increasing numbers of top players to begin to experiment with some of
TD-Gammon's unconventional plays. Here is your chance to see for yourself.
Bill Robertie played two long matches against TD-Gammon
as part of its evaluation process. "Learning from the Machine" is the complete
account of the 31 games of the first match, with annotations by Robertie.
Soft bound, 56 Pages. Level: All. Available from The
GAMMON PRESS. US$20 + Shipping
[From a flyer from the GAMMON PRESS]
[Note: This is most likely TD-GAMMON Version 1.0]
In December 1994,
The
GAMMON PRESS announced its publication of the first three (of 18) volumes
Kit Woolsey's Tournement Backgammon Series. These are text versions of Kit's
instructional Match Qiz software. There are many diagrams, which means a board
is not required to study the material. The typesetting is elegant. The comments
are elequent.
Book 1: Joe Sylvester vs. Nack Ballard 150 pages US$20
Book 2: Philip Marmorstein vs. Michael Greiner 240 pages US$25
Book 3: Mika Lidov vs. Hal Heinrich 220 pages US$25
[Authers: Oswald Jacoby and John R. Crawford]
There are many who will be quick to dismiss _The BG Book_ because it was written in 1970 and
knowledge has come very far since then.
However, I recommend this book to players who are just
beginning to study the game seriously because it offers an excellent overview of
several central features of backgammon thinking and analysis. Although the
anaysis doen't go very FAR, a reader will get clear introductions to:
- Basic probability (how dice work)
- Pip Counting (the basis for evaluating racing
chances)
- Doubling Cube theory (the 25% rule)
- Settlements (useful concept even if you never settle)
- general strategic categories (backgames, etc).
I think this book is a fine place to start, but if you
hope to get anywhere you will need to follow up with some more sophisticated
books. Magriel if you can find it, of course. Dwek's _BG for Profit_ is a good next step, too. Avoid
Barclay Cooke's _Paradoxes &
Probabilities_ and _Championship BG_,
though, because those books are wrong in their evaluation of many common
positions.
Good luck in your studies!
Albert Steg
One of my favorite books on BG is Lewis Deyong's
_Playboy's Book of Backgammon_, becasue he
alternates chapters on tactics and strategy with sections recounting various
stories from major tournaments all over the world : California, Las Vegas,
Munich, Athens, Johannesburg, Beirut.
Reading it, you get a pretty vivid picture of the BG
"scene" in the 60's and 70's. You also get some understanding of how Calcutta
Auctions work, and may learn some good anecdotal lessons about the psychology of
the game: steaming, taking insurance, etc.
-- Albert Steg
Other books seeking write-ups:
A list of
Danny Kleinman books (Backgammon)
Pages Price (US$)
----- -----------
VISION LAUGHS AT COUNTING with ADVICE TO THE DICELORN 438 $64
WONDERFUL WORLD OF BACKGAMMON 132 $18
MEANWHILE, BACK AT THE CHOUETTE 142 $20
DOUBLE-SIXES FROM THE BAR 135 $19
IS THERE LIFE AFTER BACKGAMMON? 148 $21
HOW CAN I KEEP FROM DANCING? 134 $19
THE DICE CONQUER ALL 228 $33
HOW LITTLE WE KNOW ABOUT BACKGAMMON 168 $25
THE OTHER SIDE OF MIDNIGHT 142 $20
... BUT ONLY THE HOGS WIN BACKGAMMONS 244 $37
A Backgammon Book For Gabriel. 144 $24
The Long Road To Gammon. 176 $32
[Note: Danny will be raising the price of his publications shorty (9507)]
[Original list submited by
Mika Johnsson] [sorted by date]
- Hans Berliner: ``A Program
that Plays Backgammon''
- SIGART Newsletter No. 54, October 1975
- E.B. Keeler, J. Spencer: ``Optimal doubling in Backgammon''
- Operations Research Vol. 23 No. 6, November-December
1975
- P.J. Orth: ``A Comment on
"Optimal Doubling in Backgammon''
- Operations Research 24, 1179 (1976)
- David Levner: ``Is Brute
Force Backgammon Possible ?''
- SIGART newsletter No. 58, June 1976
- N. Zadeh: ``On Doubling in
Tournament Backgammon''
- Management Science 23, 986-993 (1977)
- N. Zadeh and G.Kobliska: ``On optimal doubling in backgammon''
- Management Science 23, 853-858 (1977)
- Hans Berliner: ``Backgammon computer program beats world champion''
- Artificial intelligence 14 (1980), 205-220
- Hans Berliner: ``Computer
Backgammon''
- Scientific American 243:1, 64-72 (1980)
An on-line summary by Dr. Berliner of his work in
ai/games is found under his name at either: "http://www.cs.cmu.edu/afs/cs.cmu.edu/misc/mosaic/common/omega/web/frg.txt
http://www.cs.cmu.edu/afs/cs.cmu.edu/user/burks/frg.95
- G. Tesauro: ``Neurogammon
wins Computer Olympiad.''
- Neural computation 1, 321-323 (1989)
- G.Tesauro, T.J.Sejnowski: ``A parallel network that learns to play Backgammon''
- Artificial intelligence 39 (1989) 357-390
- G. Tesauro, ``Neurogammon:
a neural network backgammon program.''
- IJCNN Proceedings vol. III, 33-40 (1990).
- G. Tesauro, ``Practical
issues in temporal difference learning.''
- Machine Learning vol. 8, 257-277 (1992).
- G. Tesauro, ``TD-Gammon, a
self-teaching backgammon program, achieves master-level play.''
- Neural Computation, vol. 6, 215-219 (1994).
- G. Tesauro, ``Temporal
Difference Learning and TD-Gammon''
- Communications of the ACM, Vol.38, No.3, 58-68 (March
1995)
http://forum.swarthmore.edu/~jay/learn-game/systems/gammon.html
Jay Scott's backgammon in his Machine Learning in Games site.
The GAMMON PRESS (Bill Robertie)
PO Box 294
Arlington, MA
02174
U.S.A.
PHONE: (617) 641-2091
FAX: (617) 641-2660
Books, software, video, backgammon sets, precision dice, ++. The GAMMON
PRESS publishes Inside Backgammon, as well as books and booklets.
Carol Joy Cole
3003 Ridgecliffe Dr.
Flint, MI
48532
U.S.A.
(810) 232-9731.
E-mail: carlcole@sils.umich.edu
Books, software, backgammon sets, precision dice, cubes,
++. Carol Joy Cole is also the editor of the Flint Area Backgammon News.
The Backgammon Shop (Backgammon Butikken)
Gersonsvej 25
DK-2900 Hellerup
Denmark
Tel. +45 39 40 06 07
Fax + 45 39 40 01 44
E-mail: erikg@inet.uni-c.dk
Danny Kleinman
5312 1/2 Village Green
Los Angeles, CA
90016
U.S.A.
Read Kleinman, or rely on dice.
Books on Backgammon, Bridge, O'Hell, and life.
Dansk Backgammon Forlag (Danish Backgammon Press)
Gersonsvej 25
dk-2900 Hellerup
Denmark
phone: +45 39 40 06 07
or contact Erik Gravgarrd at erikg@inet.uni-c.dk
Books, sets, dice and generel backgammon equipment. You
can get all the modern books there, plus some of the old ones from the
seventies, among them `Backgammon' (Magriel). Also some software.
Crisloid INC.
P.O. Box 2205
Providence, Rhode Island
02905
tel: (401) 461-7200
fax: (401) 785-3750
They require a minimum order of $100 to obtain wholesale prices.
L. A. Strommen
6866 Meadow View Dr.
Indianapolis, IN 46226
Tel: (317) 545-0224
E-mail: diceman@indy.net
Precision Dice and JellyFish sales & support,
Backgammon Position Analyzer.
A source for hard-to-find BG books.
John C. Rather
--------------
Old & Rare Books
P.O. Box 273
Kensington, MD 20895 USA
Telephone: (301) 942-0515
John is a longtime book collector and dealer who
specializes in hard- to-find out-of-print BG books. He usually has a copy or
more of Magriel's ``Backgammon'' (aka ``The Bible''), for sale at a fair price.
A carefully annotated book price list is available upon request. John's other
book specialities are chess, magic and mountaineering.
By Site:
rec.games.backgammon
The backgammon newsgroup. Articles, problems, and information
about the game of backgammon are discussed daily in this group.
fraggel65.mdstud.chalmers.se 4321
The First Internet Backgammon Server (FIBS)
http://www.cybercom.net/~damish/backgammon/bg-faq.html
The backgammon faq.
http://www.statslab.cam.ac.uk/~sret1/backgammon/faq.html
Mirror of the backgammon FAQ in the UK. Thanks Stephen!
http://www.statslab.cam.ac.uk/~sret1/backgammon/main.html
Stephen Turner's World Wide Web backgammon page.
http://www.oslonett.no/home/swelle/NBgF.html
WWW page of the Norwegian Backgammon Federation.
http://www.cybercom.net/~damish/backgammon/nebc/nebc.html
New England Backgammon Club home page (under construction).
http://baugi.ifi.uio.no:80/~paalf/BG
Paal Fladstrups Index of Backgammon files.
http://www.ifi.uio.no/~paalf/backgammon.html
Paal's Backgammon Page.
http://www.utu.fi/harrastus/bg/
BACKGAMMONSIVU WWW page in the Finnish language
http://hdirisc9.kfk.de/www/mata/mata93/mat0m6/bg/bg.html
Marco Lau's Backgammon-Seite (German and English)
http://www.cybercom.net/~damish/backgammon/mike_quinn/fibs.htm
Michael Quinn's Guide to FIBS.
http://www.dknet.dk/~kring/backgam.html
Asger Kring's (Albatross) backgammon page.
http://www.cybercom.net/~damish/backgammon/fibshelp.html
Mark Damish's FIBS help document.
http://www.dai.ed.ac.uk/staff/personal_pages/eggertd/backgammon.html
David Eggert's page, which includes announcements and results for his
FIBS tournaments.
http://www.statslab.cam.ac.uk/~sret1/backgammon/clubs/
Stephen Turner's compilation of backgammon clubs around the world.
http://www.gammon.com
Patti Beadles backgammon page: A central repository for backgammon
related information, and Patti's personal playground.
FIBS T-shirt info.
http://www.io.org/~takeith/bg/glossary.html
Backgammon glossary.
http://www.msen.com/~lwp/BGglossary.html
Spider's BG glossary.
http://www.phil.uni-sb.de/fun/jargon/backgammon.html
Jargon File 3.0.0 - backgammon
http://sg3.organ.su.se/~tsz/equity.html
Equity Tables for different gammon rates and player strengths.
http://forum.swarthmore.edu/~jay/learn-game/systems/gammon.html
Jay Scotts machine learning in games web site.
http://www.columbia.edu/~radev/backgammon/fibsratings/
Back issues of the FIBS Rating Reports.
http://www.cs.ualberta.ca/~banks/fibs/test.html
Back issues of the FIBS Rating Reports.
http://www.yahoo.com/Recreation/Games/Board_Games/Backgammon/
A list of backgammon resources around the net.
http://www.mi.aau.dk/~mortend/elite.html
Morten Daugbjerg's homepage, which includes the bearoff program BGOUT
http://www.io.org/~takeith/bg/mpd.html
"Backgammon Match Play Doubling Strategy" By Tom Keith.
http://www.io.org/~takeith/bg/met.html
"How to Compute A Match Equity Table" By Tom Keith.
http://158.38.60.54/webcon/bgframes.html
A web page bearoff analyzer (BOA)
http://webcom.com/~markplag/backgammonpage.html
Mark's (A differant Mark) Backgammon Page. Upgrade MacFibs to include
http://www.outland.com/OutlandBackgammon.html
Outland Backgammon
http://realbeer.com/realbeer/games/beergammon.html
Games - BEERgammon
http://www.pmms.cam.ac.uk/~gjm11/programs/main.html#race3
Gareth McCaughan: programsGareth McCaughan: programs (Bearoff program)
http://www.pmms.cam.ac.uk/~gjm11/bgm/
Some Backgammon Things.
ftp://resudox.net/pub/pc/windows/games/fibsw/html/fibsw.html
FIBS/W web page.
http://este.darmstadt.gmd.de:5000/persons/fankhaus/backgammon.html
funk's Backgammon links.
ftp://ftp.netcom.com/pub/kw/kwoolsey/gammon/matchqiz/matchqiz.exe
matchqiz.exe Match Qiz demo for DOS. Self extracting.
ftp://ftp.netcom.com/pub/pa/pattib/backgammon/
kvj_*.ps.gz Kit Woolsey vs. Jeremy Bagai match. Postscript format.
Typeset by jrichter@aldebaran.cs.uni-sb.de
(Joerg Richter)
ftp://ftp.cybercom.net/pub/users/damish/backgammon/
File Name Type Notes
------------ ---- -----------------------------------------------------
00-index.txt A File that contains this message.
bg-faq-ptr A Location of the backgammon FAQ (ASCII version).
bg-faq.ascii A ASCII version of the backgammon FAQ
bg-faq.html A Hypertext version of FAQ. Read online, or save from
your browser to read offline.
bg-matches D Directory containing backgammon matches.
bg-rules.html A Backgammon Rules in formated hyper-text.
bg_1.0.tar.gz B Joerg Richter's LaTeX style & font for documenting
backgammon positions and games.
blot D A New [9511] Backgammon program for DOS.
boa_v11.exe B Bearoff program. Gives cubeless probability of
winning, and best move for up to 15 checkers on
each side in the home board. Shareware.
bqdemo.zip B Walter Trices Bearoff Quizmaster Demo. (MS-DOS)
exbgdemo.sea B Demo of Expert Backgammon for the Macintosh.
Change type to "APPL", creater to "aust"
would a .hqx (binhex) be better?
fibshelp.html A Formated HTML of FIBS help screens. AUG 94
kw_jb.tar.Z B Kit Woolsey vs. Jeremy Bagai match. Text version.
matstrat.zip B MS-DOS demo of 'Match Strategist'.
pubeval.tar.Z B Gerry Tesauros backgammon 'benchmark' function.
race.tar.Z B 2 bearoff programs: 'race2' & 'race4'. No makefile.
rfibsdos.zip B Robin Davies DOS port of rfibs and sfibs.
ftp://figment.csee.usf.edu/pub/misc/FIBS_client
tiny-fugue text interface to FIBS.
ftp://itekiris.kjemi.unit.no/pub
ftp://www.abekrd.co.uk/pub/fibs/
xfibs08 X interface to FIBS.
ftp://shuksan.cs.berkeley.edu/pub/tkfibs/
TkFibs X Tk/Tcl interface to FIBS.
ftp://resudox.net/pub/pc/windows/games/fibsw/
fibsw MS Windows interface to FIBS.
ftp://ftp.cd.chalmers.se/pub/xibc/
xibc-X.XX.tar.Z X Tcl/Tk/Expect interface to FIBS.
ftp summex-aim.stanford.edu /info-mac/game/brd/mac-fibs-10.hqx
MacFIBS Mac interface to fibs.
ftp ftp.statslab.cam.ac.uk /pub/users/sret1/backgammon/
bg2fig Board description to fig converter.
matches Backagmmon matches.
printmatch FIBS `oldmoves' to ascii bg boards converter.
psboards --> bg2fig
ftp ftp.irb.uni-hannover.de /pub/spitz/bg/
bg.tar.gz FIBS recording/playback tools.
ftp://figment.csee.usf.edu/pub/misc/FIBS_client/ms_vs_Snoopy.ascii.Z
ftp://figment.csee.usf.edu/pub/misc/FIBS_client/ms_vs_Snoopy.ps.Z
ms vs. Snoopy match annotated by Kit Woolsey, TD-Gammon, and JellyFish.
ASCII and postscript versions. The postscript version looks great!
ftp://ftp.darmstadt.gmd.de/pub/dimsys/kodim/fibstotex.tar
FIBS oldmoves to TeX converter by Peter Fankhauser.
Requires Joerg Richters bg.tex package.
ftp://ftp.darmstadt.gmd.de/pub/dimsys/kodim/rgb-positions
Peter Fankhauser's collection of problems from rec.games.backgammon.
ftp://ftp.cica.indiana.edu/pub/win3/games/bg06.zip
A weak shareware backgammon program for windows.
http://www.austin.ibm.com/pspinfo/funtdgammon.html
TD-Gammon, the program, for OS/2