Comments/Ratings for a Single Item

MT***By 'algorithmatization', do you mean finding an algorithm by which a
player can be certain not to lose? That's a good question.***

Vit: No, I mean ability for PC playing... But the question is really good
:-).


MT*** I thought at first it was obvious that no such algorithm could be
found, since Synchronous Chess is not a perfect-information game, but as I
think about it a second time, I realize it's not so obvious. But I think
it's unlikely there could be such an algorithm. Luck is a factor.***

VK. Yes: luck or telepathy  ļ.

Vit

Good afternoon, colleagues!

I very interest the basic opportunity or impossibility of
algorithmization
SC. 
In the my rules-description I put forward a hypothesis that
algorithmization of synchronous games, in particular, a chess, is
impossible. And all this time (1991-2005) I searched for the interlocutor
with whom could discuss this question. Similar, that at this forum I has
found. I very glad to meet Arnold Krasovsky here.
I give the reference to a Russian-speaking variant of the rules in the
Internet:
http://chess-e.narod.ru/Lib/sinchess.htm 

Vit

Oh, I just forgot to put my name for the last comment on the computer
game,
which I posted.

However, I wanted to make this second Post anyway, and give best
greetings
to Arnold Krasowsky. I am very pleased to read from you here. 

I am going to meet Andriy in the coming weeks, and we will have a game
then, I guess. Perhaps I even have a (good) plan for a (first)
tournament.
 

Best regards
Ralf

I am very interested in your considerations for generating a computer
implementation for the synchronuos chess game.

Well, an important point to be considered is that there does not exist
something like the a-priori best move in the synchronuous chess. It might
be possible to try to define criteria for determining the objectively
best
move in the conventional chess game. However, in synchronuous chess, the
best move of black depends on the move which white takes simultanuously.
I
thus assume, there is a fundamental difference between the conventional
and
the synchronuous chess considered from the perspective of mathematical
game
theory by Neumann and Morgenstern.

Therefore a real good computer player should have some mix between a
random algorithm and a good chess program, which determines the move.
Else, a good human player might be able to predict the computers move,
which might give him an important advantage, ... because the computer is
of course very limited in predicting the human player´s synchronuous
move.

However, I would be happy to see any computer program for the
synchronuous
chess rules. And I have the feeling that -perhaps partly ignoring the
points I just raised- it should be possible to start from an existing
computer chess game (but I am not able to do so, as I am not a
programmer)
and to modify it. I hope very much that some day a programmer will engage
in that endeavour.

It is surely a shame that ZoG development seems to be completely, totally
over.  When asked on their discussion boards if a new version is ever
forthcoming, no answer is given.  However, doing some of what Derek
suggests is redicuously difficult; it's not just a problem of limitations
in the Zillions description language that require you to have hidden pieces
and such; that view totally misses the point.  Ugliness and performance
problems are just the tip of the iceberg when addressing issues such as
Synchronous Chess.  The Alpha-Beta NegaScout algorithm that is fundamental
to most every commercial Chess program in the world cannot be used! 
Period!  If someone wants to address the Synchronous problem, and write a
program that *actually* plays this game, then they have to start from
scratch, and they do so without the benefit of any technical articles
written about the problem anywhere!  (I've read every word ever written
about writing programs for chess variants.  And, since such literature is
almost non-existant, it didn't take me long :)

That being said, although ZoG can't really be expected to play some of
the more radical games well, like Synchronous, there are some things that
its creators could do that would immediately make it infinitely better at
a large number of games, and those improvements would not require any old
ZSGs to be rewritten, and would be quite easy to implement.  They could,
for example, allow a new flag for piece types that allows the ZRF
programmer to specify the base value of a piece (excluding square
bonuses.)  The problem is solved quickly, because the responsibility is
transferred to the programmer, but it would not make any old ZRFs
obsolite, because they don't use that flag, so the program would use
default values.  And it is just so simple that it could not possibly take
more than an hour to implement.  But, it has been known that ZoG doesn't
evaluate pieces correctly for years, and there are lots of posts about it
on their discussion boards.  It seems that they are not even going to do
simple, quick improvements.  It's really too bad.

Of course, the AI of the Zillions program would perform much better and
quicker without such elaborate workarounds.  It would be nice if the
architects of the Zillions program were planning the great task of
radically revising their program from the ground-up, restructured in
light of what they know now after over 1000 games have been written for it
versus pre-1998, with more holistic adaptivity esp. with respect to game-
ending conditions, turn-orders, simultaneous moves, etc.  [Mallett once
said that an ideally written program should not nearly as often require
the use of dummy pieces. (paraphrased)]

The loss of backward compatibility for all pre-existing *.zrf's, then
necessitating that ALL Zillions games be manually rewritten to run under
the most modern version, is probably totally unacceptable to Zillions
Development and some game inventors.  Besides, I suspect Mallett & Lefler
are no longer interested in pouring monumental amounts of effort into
this minimally-profitable project.  There has not even been a minor patch
published in over 2 years.  Ed van Zon, the English-language webmaster,
is the only person who still communicates.

Mark:
Your idea is very clever, and deserves an 'A' for inginuity!  This will
allow a person to play synchronous chess against Zillions so long as
computer is White.  But there's a problem... When the computer considers
what move to play, ('thinking',) it is recursively looking at hundreds
of thousands of sequences of move-counter-move combinations to determine
which is best.  During this look-ahead, whether Zillions is White or
Black, Zillions is playing both sides and uses perfect information.  The
way Zillions decides on moves will not change from a non-synchronous game
(and I'm pretty confidant that sometimes different moves are better in 
synchronous chess; and if not, then what's the point?)  So, essentially
what you have is a way for a person to play synchronous chess with the
computer, but the computer is still just playing chess.

Derek:
It is almost certainly possible to write a program to do it... (not that I
know how to go about it...)  But the suggestion you make of a computer vs.
computer synchronous match has an additional nasty complication that is
really hard to explain, but I will give it a shot.  
To have computer vs computer synchronous, you need not one capable
program, but two seperate (and different) programs.  Here's why:  say you
try to do it with one program... You give it the ability to handle the
hidden information by not including any code that looks at variables that
it's not supposed to.  Ok, so far, so good...  So, now it must try to
'guess' what the other player is going to do.  Chess programs all do
this by assuming that the opponent will make the best move he can.  In
this case, the 'best move he can' determination is being made by the
program!  After thinking about it, the program is going to determine that
the best move is always the actual move!  So, you've slowed it all way
down by making it think about the same things over and over again, but you
haven't changed its play at all!  It's still just playing regular Chess
against itself... Wierd, huh?!?

I think you could implement something in Zillions that would work like this
game. My plan would be, program 3 players: the computer has to play first,
then the user, then a 'neutral' player. The computer's moves would take
place on an invisible 'side' board, then the human player would make a
move (not having been able to see the computer's move: you'd have to
close the panel that shows the move notations), and then the 'neutral'
player would make his move, which would always be to transfer the
computer's moves from the invisible board to the visible one. If the
transfer caused conflicts the neutral player would have to do something
complicated to resolve them. 

You could never have the computer move second, or zillions would use the
information about the human player's move.

1.  Does anyone possess a Zillions implementation (*.zrf) for this game or
any other simultaneous-move chess variant?

2.  If so, can you direct me to it?

3.  If not, is this absence due to a present limitation in the Zillions
engine of being strictly turn-based?  In other words, is it currently
impossible to implement this game for play via Zillions Of Games?

Hallo, Ralf! Thank you for your kind efforts in our Synchronous Chess! It
is true, the idea of this chess have visited my mind in the sleepless
night after an anormous doze of 'orthodoxal' chess games on the huge
park chess blackbord when I was successful with the white color and was
not with the black one. It had happend in Jurmala, Riga where I had have
a
holiday with my family in July - August of 1987 and 1988. Because of my
great duty as a depatment head of our research institute in the
tremendous
period of our country I was able to return to this idea 10 years later
when
I wrote the paper in Russian and sent it 25.04.1997 to Moscow, chess
magazine '64' (received there on 06.05.1997, no answer) as well as to
Elista town, Russia to the name of Mr. Illumzhinov, the president of FIDE
(received there on 07.05.1997, no answer). Some later the same paper in
Russian has been sent to the name of GM G. Sosonko, the Editor-ih-Cief of
the Holland magasine 'New in Chess'. I was answered they don't publish
such kind of paper.
I believe it is important now to develop the first draft of rules for
this
chess version and to start the practical games between its enthusiasts.
It is iteresting remark by Vitaly Korolev. Unfortunately, I have no any
iformation about the details of his proposal.
My e-mail address: [email protected]

Greetings, colleagues.
My name is Vitaliy Korolev (St-Peterburg, Russia).
 
Yesterday I have casually found this site and this discussion - in 15
years after creation of the game rules. It would be very interesting to
contact with Ukrainian colleagues Arnold J. Krasowsky & Andrey Krasowsky.


Concerning to the game I shall inform the following. Per 1991 I have sent
to Mr. Pritchard the rules of the “Synchronous chess” for the publication
at his Encyclopedia of the Chess Variants. He answered, that else in 1971
V.R.Patron has offered similar rules (SYNCHRONISTIC CHESS), but rules in
my edition (SYNCHRONOUS CHESS) have some of differences, and ' more
logical ' therefore have been included to the encyclopedia.

S.Y.
Vit
[email protected]

Hi Fabrice, thanks for your comment and interest for the game. The rules
invented by Vitaliy Korolev 1991 and the game implemented computationally
on the web here, differ greatly from those denoted at
http://www.hexenspiel.de/engl/synchronous-chess/

So the latter, which more or less I invented, is a new Synchronuous chess
game. As can be read on the above internet adresse, the basic idea was
from an Ukrainian fellow named Arnold Krasowsky. He developed the game
quite a long time ago, I think before the nineties. He posted the rules
to
a Russian chess Journal at that time, but never received any answer.
However, I somewhat refined the game rules, which I deemed necessary. -
so
I consider myself as coinventor of the synchronus-chess game following
the
rules described in the above web-adresse.

Concerning your second question, the rules given on the website clearly say
that when as a result of the two synchronous moves:

'b) ... a figure moves on a field, which was before the synchronous moves
occupied by a figure of the opposite color, that, however, has been
synchronously moved from this field on another field, then both figures
safely reach their new position.  Neither of the two figures is removed
from the board. Thus, in some cases the synchronous moves can result in a
mutual exchange of the positions of the two figures that have been moved
(This is possible, even if the ways of their moves overlap with each
other).' This means that a pawn that moves diagonally trying to capture a
figure which is standing on the corresponding field (!and only in this case
such a move is allowed!), simply misses that figure if it moves away in the
same moment. Thus: if 1. e2-e4 d7-d5 2. e4-d5 d5-d4 -> both pawns remain on
the board as the strike e4-d5 misses its target, as the black pawn moves
forward. It would even be possible that: 1. e2-e4 d7-d5 2. e4-d5 d5-e4 ->
here both pawns miss their target, and as a result their positions are
exchanged.

In my opinion it is absolutely essential for a synchronous chess, that it
is possible beforehand of the two synchronous moves to decide which moves
are possible, and which moves are forbidden!!! Therefore according to the
rules given on the website: A Pawn's (or any figure's) move is legal, if
it is possible according to the conventional chess rules,- i.e. assuming
that the other would not move his figures simultenously. - It is not
possible that a move is legal before the two synchronous moves, and proves
illegal afterwards.