Comments/Ratings for a Single Item

That's how Falcon Chess 8x10 operates. There is no interest in shuffling
per se. Instead from 7 to 9 arrays stand out, and they should be able to be
whittled to 4-5. Two are ''offical chess'' now, in what is still just
an intellectual construct: RNBFQKFBNR and RNFBQKBFNR. For five years
Abdul-Rahman Sibahi prefers in order RFNBQKBNFR and RBFNQKNFBR (Templars).
There are also FBRNQKNRBF (Pyramids) and FRNBQKBNRF (Cheops). On the
principle Falcon is the complement to RNB, the same way Bishop was the
complement in year 1500 to ancient Rook and Knight, there should be from
2-5 starting arrays instead of settling on only one, for our 21st-century
stamp. In all the above, Q and K are sometimes reversed but that's the same
array; and there are others named than those parenthesized. There are
different arguments for more centralized Knights and unprotected Pawns in
one or another. All these issues to resolve for line-ups of just one
''CV.'' (Carrera-Bird-Capablanca also substantially resolves to determinations of
arrays numbering 10 or 20. In general, specifying a few arrays to 4000
CVPage CVs will just increase that 4000 to 12,000 or 16,000, because most
write-ups are for one particular initial position.)

One could combine a shuffle with formations, to provide a way to have a
sound layout of pieces, but force people to adjust to openings they don't
play normally (a reason why there is shuffles).  Key would be to properly
evaluate the the strength of a formation in contrast to another one.  You
also would have to factor in things like whether a formation has castling,
can be subject to its pawns having En Passante, and so on.

Rethinking arrays, in c960 if keeping King and Queen central, there are
only rnb, rbn, brn, bnr, nrb, nbr, for 6 symmetrical, and they're all good on 8x8. Others with the Rooks' boxing in King too, with unsymmetrical included now too, number all together the 6x6, or 36, and they're all okay. 
Keeping k and q at d1 and e1 might be next in importance to what Fischer
already implemented, castling to c1 and g1. In rnbf on 8x10, there should
probably be all 24 starting arrays with k and q central, xxxxQKxxxx, but
keeping the two xxxx reflecting each other all the time, symmetrically. A couple of the 24
are not so good, such as a Bishop immediately hitting an unprotected Pawn,
but Abdul-Rahman and I had too few before with our only 5-10. 
Many c960 have Bishop (and Q) from array attacking unprotected Pawn after one pawn move, even within the 36 above.

Why not create a few standard formations, say swap the Bs & Ns, or put both
Ns on one side and both Bs on the other, that you choose randomly for each
player. Place the Ks and Qs in the middle 2 squares, again randomly for
each, short and long castling depending on each king's initial position.
So you've got RNB, RBN, RNN, and RBB, and with the K and Q random,
there's 8 possibilities/side, or 64 different opening setups. All are
close, I think, and might be balanced enough to work. If not, pick the 32
most balanced, and just use those. The generalized opening strategy for all
possibilities should be fairly obvious, but I suspect the individual
details would be sufficiently different that the memorization of the first
20 moves of a few hundred openings would be much less an advantage.
However, all of a player's actual chess ability is totally untouched by
this scheme, and the general opening principles are already very well
established. If you wish to get out into the deep end here, you could play
a game where one side had all 4 knights and the other, all 4 bishops, or
add both possibilities into the mix and strain people a little bit. ;-)

George, did we just accidentally agree on something? I didn't see your
comment before posting mine. I'd lean toward keeping the rooks in the
corners to start, anyway, so players could castle normally. Swapping the 2
similar-value pieces is relatively innocuous. In fact, now that I think of
it, it's highly likely it's been proposed before. If so, is there any
information on how well it works? Reining in Bobby Fischer - gee, that
sounds almost conservative. It may actually fly with the chess community.
:-D

Hey Joe.  I guess we could swap the positions of the Knight and Bishop (and
maybe even Rooks).  The main idea I was looking at was to keep the pieces
in their same columns, but have placement in the first three rows, with
pawns in the second or third row, and the other pieces behind them.  I
would allow for swapping of king and queen pieces.  I believe we also need
some guidelines like:
1. Castling is only permitted if the King is in the same row as a Rook at
the start of the game.  I would also look to have it so that the King would
need to have pieces between it and a Rook.
2. Rooks must be either a row behind or in the same row as the King.
3. Pawns on the second row get to move one or two spaces to start.  These
pawns that move one or two spaces are at risk of being En Passante'd. 
Pawns that start in the third row only move one space forward to start.  En
Passante is a weakness of a pawn, that another pawn can do to it.
4. Unless randomly selecting formation to start (like a shuffle), the
white player picks what formation they want to use, then the black player
selects their formation.
5. King and Queen may swap position (this changes how castling might
work).  Bishop, Knight and Rook stay in the same column they would normally
be in FIDE Chess.

My preference includes using King capture instead of checkmate, and also
promoting pawns to pieces that have been captured, but I don't want to
make this a requirement.  The focus is on formations.  Also, if people want
to have more modifications, feel free to here.  These are guidelines. 
Again, the key is adoption of the basic idea, not holding this locked down
and unchangeable.

I leave it up to others to play with more.  I have played around with
multiple formations using Fritz and ChessV and Zillions, and you get
interesting results here that look like it is playable.  Anyone want to
write this up as a legitimate chess variant?

I personally believe formations are a worthy element to be added to the
world of chess variants.  By combining it with mutators, different board,
reserve (pocket) pieces, and other things, I believe we can have a way to
have a version of chess with a LOT of different scenarios, maybe even a base version that could serve as THE main form of chess for the variant community.

Near Chess is here:
http://www.chessvariants.org/index/msdisplay.php?itemid=MSnearchess

Near vs Normal Chess is here:
http://www.chessvariants.org/index/msdisplay.php?itemid=MSnearvsnormalch


John K Lewis had suggested ideas found in Near Chess, but then chopped off
the ends that were vacated, and was the first run at Simplified Chess.  The board then became 8x7 and the final version of Simplified. I believe a Simplified Chess board is a good addition to the world of Chess variants.  I did disagree with Mr. Lewis chopping off the last row in the initial Simplified Chess.

An entry as a 'variant' on the CV site, is found here:
http://www.chessvariants.org/index/msdisplay.php?itemid=MSmultipleformat

Please continue discussions there.