Comments by AntoineFourriere

Unfortunately, you can call a draw at PMC only after 50 moves without
capture or promotion, so maybe one or both players are simply taking time
to think. Still, a general rule that third repetition is a draw unless the
game rules specify it to be a loss or a win might be welcome for GCT #2.
Anyway, I'm all for starting Round 3.

The game looks quite interesting.
A couple of remarks:
The Bishop is weaker than the Knight even on the 49-square center (a
Bishop cannot slide more than six squares in any direction, and a Knight
isn't at all hampered by the limits of the board). A Chancellor is also
much stronger than a Queen. Maybe the Terrain should be a bit more
Bishop-friendly, but I don't really see how.
The Serpent is a very nice piece, and I would suggest to disallow the
capture of a Serpent by a Serpent to keep them longer into play.

At both Chess and Alice Chess, the Bishops are restricted to one half of
the squares. But at Alice Chess, this holds true also for the Knights, and
for the Pawns once they've completed their first move.

So you can paint the squares in eight different colors, each color
meaning:
This square will not accept:
1) the dark-square Bishops (OR the light-square Bishops)
2) the Knights which started on a dark square (OR the Knights which
started on a light square)
3) the white Pawns whose first advance was of two squares and the black
Pawns whose first advance was of one square (OR the white Pawns whose
first advance was of one square and the black Pawns whose first advance
was of two squares)

This amounts to eight different square types.
Something like:

   1 2 1 2 1 2 1 2    3 4 3 4 3 4 3 4
   5 6 5 6 5 6 5 6    7 8 7 8 7 8 7 8
   1 2 1 2 1 2 1 2    3 4 3 4 3 4 3 4
   5 6 5 6 5 6 5 6    7 8 7 8 7 8 7 8
   1 2 1 2 1 2 1 2    3 4 3 4 3 4 3 4
   5 6 5 6 5 6 5 6    7 8 7 8 7 8 7 8
   1 2 1 2 1 2 1 2    3 4 3 4 3 4 3 4
   5 6 5 6 5 6 5 6    7 8 7 8 7 8 7 8

Or you can have three ways of painting the squares:
light and dark as usual, in two colors I'm referring to as B-colors, to
separate the Bishops in two classes (1368 vs 2457)
reversing the colors of one chessboard, in two N-colors, to separate the
Knights in two classes (1467 vs 2358)
even-numbered rows of one chessboard and odd-numbered rows of the other
chessboard, and vice versa, in two P-colors, to separate the Pawns in two
classes (1278 vs 3456)
However, a Bishop can be captured only by a Bishop of the same B-color,
while a Knight can be captured only by a Knight of the other N-color and a
Pawn only by a Pawn of the other P-color.

Thanks again for the congratulations, Roberto, but there is little doubt
that having one hour for each move reflected in the outcome, especially
when I ended up sacrificing my Rook in our game of Anti-King Chess. Most
participants (and non-participants) had several other things to do, such
as work, family, studying or maintaining this site. 
In my view, the only fair equalizing methods are
1) to give all players enough time to think between each move, that is,
allowing them a pace of only one move per week at times in a given game.
For next year, I would suggest a January to June round-robin of six or
eight games, and a seven-player September to November final of six games
because these seven players might have enough time. (I would also allow
two players to replace their assigned game with any game that has been
played in any yearly tournament such as this one or last year's
tournament.) And I think you need a more lenient pace at the beginning of
the games, not merely to avoid blunders, but also to assess the possible
strategies, and because after twenty moves, half the games are already
more or less decided and you can drop them anyway if you're losing.
2) to play each game in four hours, or by slices of ten moves by hour with
one player playing a secret move at the end of the slice, like in
FIDE-Chess. Of course you need to have both players connected at the same
time (probably feasible on Saturdays and Sundays), but there is also the
problem of your Internet connection. It is one thing to lose an
independent game because of a technical problem, but I wouldn't like to
lose a Tournament that way. (Now, the penalty for not playing within the
clock in the middle of a ten-move slice in a given game could be the piece
of your choice (unless your opponent doesn't want it) the first time, that
same piece to drop for your opponent the second time and loss of the match
the third time. But it sounds overly complicated.)
Nevertheless, the implementation of that kind of device might be useful.

You can edit the preset and create a new one (different colors, different
numbers of files/rows, different piece locations, another set of
pieces...). Then you must update the new preset and paste its new FORM
block on a web page on your own hard disk to issue an invitation.
I think you should use a slightly different name if you do not simply
change the colors, like Viking Chess on 108 squares as opposed to Viking
Chess. Otherwise your log will end up in the original directory.

Même si la Dame adverse est clouée, elle peut encore prendre votre Roi
avant que vous ne preniez le Roi adverse.
Donc vous ne pouvez jamais mettre votre Roi en échec.

There is an inconsistency between David's animated illustration for the
Chameleon and Peter's Zillions file.
Is a Chameleon allowed to capture a Long-Leaper and/or a Withdrawer and/or
an Advancer and swap with an enemy Swapper in the same move, as shown on
the animated illustration but not allowed by the zrf?
Besides, is a Chameleon allowed to swap with its own Swapper?
(The zrf allows it, though it is not formally equivalent to the Swapper
swapping with its own Chameleon, since the Swapper may be immobilized.
For the record, a similar rule would be unplayable for a variant with
Chameleons and Shields. The Chameleon and Shield would protect each other,
and the Shield would protect the King.)
If the answer to both questions is yes, is a Chameleon allowed to capture
a Long-Leaper and/or a Withdrawer and/or an Advancer and swap with a
friendly Swapper in the same move?

White/Black piece on e1 is immobilized and commits suicide: @-e1
(the preset wouldn't accept it as move 1, because of comparisons between
this move's parameters and last move's parameters, but Rococo features
no suicide at move 1 anyway)

White Swapper on e1 moves to unoccupied square e2: S e1-e2
White Swapper on e1 trades positions with White/Black piece on e2: S
e1-e2
White Swapper on e1 explodes with Black piece on e2: S e1-e2; @-e2

White Archer on e1 moves to unoccupied square e2: B e1-e2
White Archer on e1 kills Black piece on e2: B e1-e2


I chose B/b (Bowman) for Archers (and later D/d for Birds) and read
Peter's comment only after Michael and Roberto started a game of Fugue,
so I haven't corrected that inconsistency.

Regarding game creation, I think I am of both schools. I tried firstly
with Bilateral Chess (obviously flawed), then with Chess on a Larger
Board with not so few pieces dropped (perhaps too unbalanced) to come
up with a satisfactory extention for Chess, but the games I am most
happy with came to my mind randomly, usually out of an external contraint
such as the number of squares (Jacks and Witches, Pocket Polypiece Chess)
or Roberto's idea of a game without capture (Bifocal Chess).

I think we should use Game Courier to revive Glenn Overby's
Invent-and-Play formula on a yearly basis, with no other time limit.
Games which are to a large extent the juxtaposition of a number of
somewhat contrived pieces, or which threaten to last for about a hundred
moves, like Achernar/Deneb, Chess on a Larger Board with not so few
pieces dropped, Optima, Heroes Hexagonal Chess or Chess with Terrain
should start there rather than on a Game Courier Tournament.

I would suggest to allow comments on a position, either from the players
or from the kibitzes, only five moves later. Otherwise it will taint the
game needlessly.

The World against Kasparov is also a nice idea, however I think we should
enrol several Kasparovs, again on a yearly basis, because the Kasparovs
may be unavailable at times. On the other hand, the World should have a
definite answer period, say three days for choosing their move by a poll,
with ties broken randomly, which means Fergus would have to implement
such a device. Maybe a higher Kasparov in the GC Tournament should not
participate against a lower Kasparov.

Thanks to Fergus for a well thought-out Tournament, and for having devised
Game Courier.

Thanks to Michael and Carlos for their congratulations.

Actually, although I won three of them, my games of Pocket Mutation Chess
(my first choice) against Carlos, of Anti-King Chess (my second choice)
against Roberto, of Alice Chess (my third or fourth choice) against Fergus
and particularly of Chessgi against Michael Madsen were all far from easy,
and could each have ended into a defeat.

As for blunders, I doubt there will be many of them in next year's
Tournament, at least from the players who participated this year.

How about five moves without computer analysis or ten moves with computer
analysis? Let the kibitzers take some deep thinking by themselves too!

I have written a comprehensive postauto1 and postauto2 code (with perhaps a
few bugs) which covers the 16 possibilities for Chess With Different
Armies. (All there is to change in each preset is the title in 'game',
the initial setup in 'code' and two arrays for possible promotions in
'pregame'.) Would it be possible to refer to a common postauto1 entry and
a common postauto2 entry, instead of having to maintain 16 identical
postauto1 entries and 16 identical postauto2 entries?
It would also be useful when there are several presets for a same game.

I have replaced the Lion and the Deer by the Elephant and the Man in
Galactic-1 (I don't know if it impacts other games), but it seems to have
no effect on transparency.
Why is there only one of the two Black Flyer-Elephants -- whether
alfaerie/elephant.gif, galactic/elephant.gif or galactic/lion.gif -- which
is transparent?

I also wish to compete in this Tournament. Please count me in.

Regarding time limits, I have no strict formula in head, but I'm on the
side of leniency. Yes, 14 days of spare time does seem an improvement.
I'm also in favor of giving the players the possibility to allow each
other more time through a command (rather through repeating idle moves).
I would suggest to state that the referee may start a new round of games
without the agreement of the players which haven't completed all their
previous games (but not without two-week advance notice).

Another problem might be: what if there are more participants than last
time? How do you decide who is going to face whom? Through a random draw
of the opponents (alphabetical order is as good as any the first time, but
might become bothersome the following year), and game assignment only
later? (Perhaps Fergus might then propose an exchange of opponents to a
group of four players in order to meet their preferences.) If and when
such a problem occurs (and I hope it eventually does), I would suggest
some kind of play-off (say, a round of four/six games between the top
five/seven, with ties broken through the round-robin ranking) at the end of
the year to mitigate the luck factor.