Comments/Ratings for a Single Item

This is the obvious place to answer Mats questions/objections from the
following post on the game Cataclysm:
'I see no point in this, as this variant is virtually unplayable. The
tactical capacity of short-range-pieces is such that this takes too long
to play. Could somebody please explain the credo behind these constructs?
Are they to be regarded pieces of art, or what? Why not settle for more
modest constructs?
/Mats'
On the CV wiki, there is a discussion [aka: argument] on the definition of
chess. It's not entirely serious, and seems to indicate that, in defining
'chess', we are actually stating our own views of what chess is/could
be/should be. We apparently define ourselves more than chess. You see
chess, or chess variants to be more precise, as something to be played on
a small board with about 15-20 pieces per side, if I understand you
correctly. What would you consider the size limits of playability, and
why?

One size I've found very interesting is the 12x16 board - don't know just
if this size fits here, but with a little adjustment to 1 game, I will have
done 3 12x16 games using 2 chess-sets worth of pieces. All 3 games can be
played with 2 standard chess sets and 3 boards, one of which is cut in
half. There are a number of other 12x16s out there. Some, like Cataclysm,
are quite good.  But very large, good games can be hard to find, so I like
the idea of a 'Very-Large' thread. So, my only question at this point is:
is the 12x16 a very-large or a large? [Okay, it's a transition size, and
might deserve its own thread. I'll also offer another oddball, a 9x21,
and ask about that.] 
Good idea. I'll try to come up with something worth adding.

Might I suggest the following system for classifying the size of a variant?

Where the upper limits are based on the simple formula
(10 x 2^{(size category)}) -1

I disagree. This system can only be applied to 2D square games.

3D, 4D and hexagonal games shouldn't be classified like this. For
example, Raumschach has 125 squares (cubes,) which, according to your
system, is large. But in fact it's a very small board. Joe Joyce's 4D
Hyperchess has 256 squares (cells, tessaracts,) but it's really small.
[Consider that the restricted king of Hyperchess can go from one corner to
the other in 6 moves. A full 4D king which moves orthogonally, diagonally,
triagonally, and tetragonally can go the same distance in 3 moves.]

And of course, you're the Hex Chess expert on the site, you wouldn't
call the 91 cells of Glinski's Chess 'large'.

I think the size of the board is better classified in relation to the
speed of the king. If the king can cross from corner to corner in, say, 3
moves, it's a small board. And so on.

Just a little something to add to this - the 4x4x4x4 4D board is 16x16 when
laid out in 2D to actually play on. Yet, because the movement rules are
written for a 4x4x4x4, the king, or any other piece, *cannot* move more
than 3 squares in any one 'direction'. 
I do agree that 'size' is related to dimensionality. A 4x4x4x4 4D board
is the smallest really useful 4D board. A 3x3x3x3 needs gimmicks to work
well, or at all, and isn't big enough for pieces to have any scope for
their moves. How does a knight move on a 3x3x3x3? Put it in the middle,
and it has no moves at all. The same sorts or arguments can be made for
3D, where 4x4x4 is tiny. I would argue that an 8x8 in 2D is also a small
board, though, so you'd want to take this discussion with a grain of
salt.
Possibly a better criteria would be what people will actually play. Given
recent history, the 12x16 is the very largest that most people will try
now. For those of us who like the larger sizes, the trick is to come up
with enough good games to bring the others along.
This leads us to the question: What makes a large game interesting? Just
what can you do on a large board that you can't do on a small one?

You are quite correct, this formula applies only to 2D-square boards. I have tried to generalize it here

A result of this suggested classifcation is an assignment of an objective size-category or size-index number. The descriptions, on the other hand are subjective. Thus an 8x8-2d-squared-cell board is classified as a category-3 sized board - whether you want to describe category-3 boards as small, standard, or maybe even glè mhòr¹ is entirely a matter of persional preference.

1. very big

After looking at Graeme's size numbers here and his extension of the
system in the CVwiki, I have to agree this is a pretty fair system, and,
to me anyway, feels right. Good job.

See Large,Hexagonal for a list of 15 Large Variants with hexagons, including Fergus Duniho's Hex Shogi 81. The 'Large' label may have been inconsistently applied in the past. Getting back to my original proposal for a tag (or whatever the computerspeak is), I assume that both 'Large board' and 'Very Large board' can be clicked, if the inventor wishes to be listed in both categories. Note that a board 15 hexagons in diameter has 'only' 6x(1+2+3+4+5+6+7) + 1 = 169 cells, so it does not meet any of my conditions for a 'Very Large board'. The same holds true for a 3D variant on a 6x6x6 board.

INFINITE CHESS VARIANTS seem to be have been filed under 'Different shaped boards or squares'. I regard them as 'Very Large boards'. The reason we assign labels is to help both regular and new users search this site, using tools like Chess Variants Query. And Joe, '4' is a very large number of dimensions :>)

While I'm waiting for parts to fix my computer, I sometimes get to borrow
an old laptop. Figure I've been too quiet lately, and George has been
talking baseball, so I say: 'Go Mets!' and point out he's still talking
little boards.

A while back, I argued that it's possible to design a 100x100 [10,000
squares] chess variant that is both interesting and playable by human
beings in a reasonable amount of time. At that time, I had a very playable
16x12 [Chieftain Chess], and was experimenting with a 24x24 and a 30x20.
The experiments were fruitful, but possibly* a little too ambitious - not
in size, but in scope, as they had too many ideas crammed in, far more
than a new design should have. I plan to re-visit them in the future, and
pare them down to manageable games. Not make them smaller, but rather
sleeker and cleaner. [In other words, dump all that extra stuff I crammed
into them... maybe someday I'll learn.] Anyway, went back to Chieftain,
and realized it was scalable. Extend the board to 24x12, play a game, and
it becomes obvious that this game can literally be extended indefinitely,
without any increase in the number of turns per game, which is maybe 40.
Of course, the games are multi-movers, with 1 piece in 8 able to move at
start. And now, even though I'm a Mets fan, I'll take a leaf from the
Falcons playbook, ending here, to continue later.
* 'possibly', ha! absolutely is more like it.

In Pritchard 'ECV' is Charles Fort's chess on 1000 squares. In CVPage
David Howe has Infinite Recursion Chess and Mega-Chess. A. Missoum has
Geometric Sequence of Chess Games.  Kieth Douglas has Chess with an
Infinite Board, Tim Converse Infinite Chess, T. Thomas another one called the same Infinite Chess.

On the other hand, it's possible I'm a Giants fan. The big guys can
surprise you. You expect them to be bad, they start off that way, and
bang! suddenly they win the Super Bowl. 

What are the features that cause western chess to favor white a bit? I
propose 2 reasons. First, the board is [very] small. Edge effects alone
cripple the knight and bishop on an 8x8 board. Whoever can grab more
maneuvering room is favored here, especially considering the nature of the
western chesspieces. And 1 square on an 8x8 is almost 2% of the entire
board. It's not surprising that the one who moves first gains a slight
advantage. [Given the geometries involved, it would be more surprising if
that didn't happen.] Second, the pawn is a directional piece on a small
board. Every pawn move made is irrevocable. And the pawns must move
directly toward danger, on this small board. After one or two moves, the
pawn has a high chance of being captured. Limited mobility, instant
contact with the enemy, these 2 things favor the first to move. [Wonder
what a pawns only game would play like? Set them up on the 2nd and 7th
rows, use double first move and en passant, allow queening, and see who
wins by eliminating the other side. Or force 1st promotion to king, then
allow non-king promos after.]

Dale Holmes' Salmon P. Chess has 7500 squares, and the write-up alone is
worth reading. Dale also did Taiga, a full 10,000 squares, with rules
found on the CVwiki. Sadly, the fine diagram he provided is gone, victim
of a broken link.

Why larger games? [Because I've done a nice 30x15 that I want to push?!?]
Because you can do things in them that you cannot in the small games. A
good large multi-move game allows concentrations of strength in time and
space that cannot be duplicated sequentially or on smaller boards. It
gives the opportunity for new strategies and tactics, and radically
different setups. It allows you to play games that are recognizably chess
variants, but ones that use different organizing principles to achieve
some unexpected results also. And because the game I'm working from is
scalable, it can be clearly demonstrated that there are effects that only
occur on larger scales. Besides running a 16x12 up to 24x12 and 30x15, I
also cut it down to 8x12 and 10x10. These 2 small games start each player
out with 2 moves/turn for 16 pieces/side. The 16x12 has 4 moves/turn and
32 pieces/side, the 24x12, 6 and 48, and the 30x15 8 moves per turn and 64
pieces per side. They all end in some 35-40 turns generally. The pieces in
each game are identical, with a minor difference in one game. So the
variations of strategy and tactics that occur are easily related to the
change in scale.