Site review - History of computer chess programs

---

Earlier this year I realized that my master list of chess history
links was missing a site for computer chess. Given that the Internet
is the meeting place of the computer literate, you would think that
there would be an abundance of choice, but I found only a few sites
that covered the topic in any depth. I finally decided to include...

History of computer chess programs by John Marountas
http://www.digichess.gr/home.php?lang=en&class_ID=1&choice_ID=6&q=1

...which is based on...

Computer Chess History by Bill Wall
http://www.geocities.com/SiliconValley/Lab/7378/comphis.htm

...plus some additional material. Another very good link is...

A short history of computer chess by Frederic Friedel
http://www.chessbase.com/columns/column.asp?pid=102

...by the man who has helped make computer chess history for almost
two decades. Another good choice is..

Schach & Computer : Eine perfekte Kombination! by Michael Scheidl
http://www.computerschach.de/einleit/start.htm

...a German language site which may be the most comprehensive set of
pages on the subject. I should also add it to my list, but haven't
yet.

Who is John Marountas? An earlier (1999) site...

'Are you ready to travel to my Island?'
http://www.agriroot.aua.gr/micisland/

...by Giannis Marountas, where 'Giannis' must be the Greek form
of 'John', tells us that his interests are computer programming and
agriculture, as well as chess.

The 'History of computer chess programs' is a timeline split by
decade into six pieces, where the first piece covers history up to
1959. The six pieces together make an essay of about 3000 words and
sit in the left sidebar menu under the [Information] topic. Other
main topics are [Homepage], [PlayZone], [Beginners], [Articles],
[Publications], [Small PC], [Algorithms], [Database], and [Publish].
Each main topic has a handful of second level topics.

'Welcome' under [Homepage] informs us that the full name of the site
is Digital Chess Network and that 'Digichess (derived from Digital
Chess) is an online service, offering computer chess news from
anywhere on the globe'. 'What's new', another [Homepage] subtopic,
tells us that there are 'New EPD Tests!' and that a 'New Chess
Server!' opened on the site in August.

The chess server, under the [PlayZone] topic, seems to offer the type
of service which keeps track of moves between two players who are not
necessarily logged in at the same time. This is an improvement on
email chess, because the server can detect illegal moves before they
are sent and can tell when a move has been made by one opponent and
when it has been retrieved by the other. You don't need to register
with the service to follow the moves and comments in 'Classic Chess'
(43 games in progress) or in 'Fischer Chess' (6 games).

Along with the essay on computer chess history ('History of CC'), the
[Information] topic lists 'Past events', 'Com Vs Hum' (Computer Vs
Human), 'CC Calendar', 'Ratings', and 'Newsletter'. 'Past events'
lists eight events in random order with crosstables; the earliest
is '1981 - 1st (Official) Dutch Computer Chess Championship' and the
two most recent are from 1997. 'Com Vs Hum' covers six events from
the 1st Harvard Cup (1989) through Van Wely vs. Rebel (2002). One
event listed as '???? - Fritz and the Grandmasters!' was in fact a
1995 event at Godesberg, Germany, where Fritz 3 achieved an IM
norm. 'Ratings' has six links to lists by various computer rating
services, including the well known Svenska Schackdatorforeningen
(SSDF). 'Newsletter' lists five published by Digichess, which I'll
discuss later.

The [Beginners] topic answers questions like 'Where to find source
code from other authors to study?' and 'What means MinMax?', which
should be 'MiniMax'. One question which interested me was 'What is
the ".eng" format?' This is answered, 'the format of the Chessbase
engine files. For more details go to the Chessbase website :
www.chessbase.com.' A search on that Chessbase site proved fruitless,
but a Google search pointed me to a chessbase.de page that explained
the term. It's the piece of chess software that generates and chooses
moves, as opposed to the piece that serves as the user interface.

The [Articles] topic has a 'Guides' subtopic with two links, one on
Winboard/Xboard and another on the 'MinMax' algorithm; a 'Reviews'
subtopic which is empty; 'Interviews'; and 'Tourneys' with articles
on computer events. I found 'Interviews' to be one of the most
interesting sections on the entire site. The first is a recent
interview with Amir Ban, who is half of the team (with Shay
Bushinsky) behind Deep Junior, which won the 10th World Computer
Chess Championship in Maastricht in July. With the exception of a
2001 Kramnik interview, apparently extracted from elsewhere, the
other interviews, all from 2002, are with developers programming for
handheld devices.

One interview is with Richard Lang whose Mephisto programs won six
consecutive World Microcomputer Championships from 1985 to 1990. Lang
has a site at...

ChessGenius
http://www.chessgenius.com/

...with details on ChessGenius for three different platforms. In the
interview he says, 'A friend told me about the Palm and that it had
the same processor as the Mephisto chess computers that I worked on
in the past and I bought one. At the time the only Palm chess
programs were really weak, but playing chess on the Palm was great.
Making moves with a stylus is much easier than with a mouse, there
are no pieces to lose as with pocket chess computers and of course
the portability. [...] I decided to have a go at porting ChessGenius.
I used the world champion "Roma program" as the engine and wrote a
new user interface.'

Another interview is with the team of Frank Schneider and Kai
Skibbe...

PocketGrandmaster
http://www.pocketgrandmaster.com/

...One of them says, 'At the WMCCC in August 2001 it was the first
time I had a Pocket PC in my hand. The bright display of the iPAQ was
much better than that of the Palm IIIx I owned and I was really
exited how much fun chess was on the Pocket PC. For the first time in
many years I really enjoyed playing a computer!'

A third interview is with Christopher Theron...

Chess Tiger
http://www.rebel.nl/chesstiger/

...who says, 'I really love the Palm and I was excited about the idea
of being able to take my chess engine with me everywhere I go. Since
many years I have been looking for a computer that I could carry with
me most of the time, and that would help me to manage information
efficiently.'

The interviews were all conducted by Pavel Mahamood, who has his own
site...

Chess-0-Mania
http://members.bellatlantic.net/~vze33v6m/

...covering computer tournaments that he has directed himself.

As I was reading the interviews, it occurred to me that it's just a
matter of time before I'll have to create a new page for these
devices on my World Championship site. That time may come very soon!
Another Digichess page announces the '1st Computer Chess Handheld
Tournament (CCHT), organized by Digital Chess Network' to be held 2-3
& 9-10 November and cosponsored by Chess Informant.

The remaining topics are [Publications], which lists six articles (of
which only two are available) and links to Amazon for books; [Small
PC], which should cover handheld devices but is empty except for a
list of about 20 links; [Algorithms], also largely empty; [Database],
with games from ~45 events, descriptions of software (programs,
utilities, and fonts), EPD ('Extended Position Description' in case
you were wondering) tests, and links ('Sorry !! This service is
currently unavailable !!'); and [Publish], to exchange links with
Digichess in various ways, where 'Top CC Sites' is also 'currently
unavailable'.

Along with the main topics that I've just outlined, there are a few
other navigational possibilities. 'The ChessExpress Webring' includes
more than sites related to chess computers, but at least one...

The ChessBrain Network
http://www.chessbrain.net/

...could be the mother of all these sites -- 'The ChessBrain project
seeks to create a massive chess playing computer by utilizing the
idle processing power of networked machines.' 'Hosted sites', on the
right sidebar of each page introduces four new topics...

Pyotr Engine by John Marountas : 'Pyotr is a program that knows all
rules of the ancient game of chess and is programmed to keep you
company at a game of chess!'

Infinite Loop by Igor Gorelikov : 'The main goal of Infinite Loop is
to rate all free chess engines at three fixed time controls on the
same hardware.'

Olympiad : Computer Chess Olympiads History, since 1989.

CCHT : the 1st Computer Chess Handheld Tournament that I mentioned
earlier

...The CCHT page also announces, 'This event is our first fruit of
hard work after one year of existence! Yes we have our birthday at
1.9.2002!', so congratulations are in order!

What about the newsletters? There are five of them -- News (about
Digichess), Handhelds News, Update your engines, Commercial Chess,
and Recommended Books. Unfortunately, all of them are listed with
publication schedules that have never been respected. 'Recommended
Books', for example, is 'published every month', but appeared once,
in June.

In that issue, the 'Book of the month' was 'The System : A World
Champion's Approach to Chess' by Hans Berliner. The review, which is
three paragraphs, mentions that 'his Hitect program won the World
Computer Championship twice'. In fact, that should be 'Hitech', and
in spite of Dr. Berliner's tremendous accomplishments, his program
never succeeded in winning the world championship. My own page...

World Chess Championship : Computer Chess
http://www.mark-weeks.com/chess/wcc-comp.htm

...tells me that Hitech tied for:-
- 1st-4th, Cologne, 1986; Cray Blitz won on tiebreak
- 3rd-5th, Edmonton, 1989; Deep Thought won 5-0
- 6th-9th, Madrid, 1992; Chessmachine 1st
- 6th-10th, Hong Kong, 1995; Fritz won on tiebreak
which means that Hitech tied for first once, but didn't win the
title, losing to Cray Blitz in the last round.

All of this -- the empty topics listed in the menus, the newsletters
on imposed schedules which are not met, the factual errors -- gives a
mixed impression of the Digichess site. The content is certainly
promising, but the objectives are higher than the available resources
are able to attain.

The site also has a problem spawning browser windows. Click on a link
and the page opens in a new window. Click a link on the new page and
another window opens. For links which appear to have the same
functionality, some open new windows, some don't. Considering the
nature of the site, my guess is that most visitors are sufficiently
computer savvy to open a page in a new window when appropriate or to
return to a previous page when appropriate. The spawned pages also
have inconsistent attributes. I wanted to search one long page for a
name but was unable to invoke the search function because the
toolbars were all missing.

That's enough nitpicking. Let's take a brief look at the essay on
computer chess history.

---

As I mentioned earlier, the essay 'History of computer chess
programs' uses a lot of Bill Wall's material. Where Wall's page
starts, 'In 1947, Alan Turing specified the first chess program for
chess', the Digichess page starts with a paragraph on von Kempelen,
another on Konrad Zuse, and continues '1947 : Alan Turing specified
the first chess program for chess.' It's not clear why von Kempelen,
whose Turk was a clever magic trick, is mentioned, but Torres y
Quevedo, who built a real machine to play King & Rook vs. King
endgames, isn't.

Wall's history ends with Kasparov's loss to Deep Blue in 1997,
although there's an additional paragraph on the highest rated
computers, according to the SSDF, at the beginning of 1999. The
Digichess essay ends its page covering the 1990s with a word-for-word
copy of Wall's final two paragraphs, then lists exactly three 2002
events on its page covering the 2000s. The years 2000 and 2001 are
missing completely.

After reading this essay, I decided that computer chess history can
be broken into three phases. The first phase was when machines played
a weak game. It started when two of the founding fathers of computer
science -- von Neumann and Turing -- used chess to study how
intellectual activity could be tackled by a machine. Shannon's
paper 'Programming a computer for playing chess', published in 1950,
was the first to describe the programming task. Progress was slow but
steady in the 1950s & 1960s, and the first computer tournament took
place in 1970.

The second phase was in the 1970s and 1980s, when computers continued
to beat better and better players. The Soviet computer KAISSA won the
world championship in 1974. American machines won the next five title
events, with Deep Thought winning the last American title in 1989.
The American team of Dan and Kathe Spracklen won the first four world
microcomputer championships, but they were eventually eclipsed by the
Briton Richard Lang, who won the next six. David Levy made his famous
ten-year bet in 1968, won in 1978, and finally lost to Deep Thought
in 1989. Since the early 1990s, the difference in strength between
computer and microcomputer has gradually disappeared.

The third phase would be the current rivalry between the best human
players and the best machines, starting with Deep Thought and its
successor Deep Blue. As practically everyone on the planet knows,
IBM's beast beat World Champion Kasparov in 1997, and then
disappeared from the chess scene. The Kramnik - Fritz and Kasparov -
Junior matches next month will tell us where we are. Not many people
expect the machines to hold their own.

The fourth phase will be when the best human players have no chance
to win against the best machines. There may come a day when the best
humans won't even be able to draw.

One problem with computer chess history is the nature of general
chess history, which centers on the best players. There have been
plenty of books on the different world champions, but very few on the
world championships. Many amateur players know something about
Capablanca and Alekhine, but very little about New York 1924 or
London 1927. People and personalities are simply more interesting
than round-by-round accounts and crosstables.

With computer chess history, the players' personalities are
nonexistent. The successive versions of a specific program get new
numbers (2.0 -> 3.0 -> ...) or are renamed completely (Chiptest ->
Deep Thought -> Deep Blue). You might expect the developers to become
the center of attention, but they don't seem to be interested in this
role.

Along with machines and developers, some other topics for historical
research would be competitions (computer vs. computer, vs. human),
manufacturers, algorithms, and databases (like ChessBase, or for
retrograde analysis). I have more to say about this subject, but I'll
save that for a future issue of Chess History on the Web.

Bye for now,
Mark Weeks

---

[Send comments to...

Chess_History-owner@yahoogroups.com

...The bookmarks for the sites reviewed here may be found at...

Chess History on the Web : Recommended sites
http://mark_weeks.tripod.com/Chs-hist/chs-hist.htm

...The list archive is at...

http://groups.yahoo.com/group/Chess_History/messages

...and contains copies of previous reviews. To subscribe:-
- by email, send a message to
Chess_History-subscribe@yahoogroups.com, -or-
- by the web, go to http://groups.yahoo.com/group/Chess_History/join

All subscriptions are confirmed with a verification message.]