Check out Janggi (Korean Chess), our featured variant for December, 2024.


[ Help | Earliest Comments | Latest Comments ]
[ List All Subjects of Discussion | Create New Subject of Discussion ]
[ List Earliest Comments Only For Pages | Games | Rated Pages | Rated Games | Subjects of Discussion ]

Comments/Ratings for a Single Item

Earlier Reverse Order Later
Generic Chess Piece Creation System. This is a system for construction of pieces, using ideas from RPG games.[All Comments] [Add Comment or Rating]
Eco wrote on Sun, Jul 20, 2003 04:43 AM UTC:Excellent ★★★★★
I really like this one! But there are some things that need clarifying.

First off, how does the reproduction ability work? There's no explanation
of that other than 'spawns pieces.'

Also, is there any convention for pieces that can move in a particular
direction, say 3 spaces, but ONLY that many spaces?

Anonymous wrote on Mon, Oct 25, 2004 08:30 PM UTC:
How much does a piece like a Cannon (slides orthogonally, leaps over 1 and only 1 piece, then lands anywhere beyond) cost?

Corgi wrote on Fri, Dec 24, 2004 11:49 PM UTC:Excellent ★★★★★
I really like this!  I'm taking a whack at a chess variant themed off a
fiction series, and this really helps me figure out the value of the
pieces for balance.

However, two things I spotted that I think are errors - 'K' is being
used as the symbol/abbreviation for both Knight and King; and the King
(Radius 1) is valued at 20 in one spot then 30 in another.

In addition, shouldn't the basic King be worth more?  (3*4) for
orthogonal, (2*4) for diagonal = 12+8=20... then (2*2)/4 for the castling
move = 1... then (21*2)*5 for a Royal piece which can only be captured by
checkmate.  That's... [sheepishly pulls out calculator] 210.

Very practical page - thanks for sharing it.

Zaratustra wrote on Sun, Dec 26, 2004 11:47 PM UTC:
Do feel free to add any modifiers that you haven't noticed here and you
believe sensible. These values are hardly set on stone.

As for 'spawn pieces', I intended it to be 'add a piece of the same
type on any square the piece can move to'. More complicated constructs of
this type may require additional modifiers.

Andy Maxson wrote on Wed, Feb 28, 2007 05:16 AM UTC:Good ★★★★
the elephant alfil plus dabbaba is actually 1/4 bound. and isn't the zebra and antelope weaker than the knight? they are more akward. how about a hopper category? lot's of these pieces come from works of lim ther peng or adrian king

David Paulowich wrote on Wed, Feb 28, 2007 02:52 PM UTC:
P=15, Ferz=25, N=45, B=45, Commoner=60, R=75, Q=135 are the numbers I have worked out over the years. See my Shatranj Kamil (64) page for an early listing of these (endgame) values.

I can accept the values of Wazir=22 and Alibaba(DA)=32. Andy Maxson pointed out that the Alibaba (Elephant on this page) does not have a value of 37. Ralph Betza has an index page here for his work on different chesspieces. The WF (Commoner) is tricky to evaluate, as its strength varies during the game. My philosophy is: eventually you are going to have to play an endgame, so that is where the 'true value' of each piece is determined. Regarding the DA (Daffy or Alibaba or Elephant), Betza believes that the DA is noticeably weaker than the N in practice. Elsewhere he has stated that the practical value of a DA-rider piece seems to be a bit less than a Knight. In other words, changing a weak piece (limited to one quarter of the board) to a rider does not seem to add much value.


David Paulowich wrote on Wed, Feb 28, 2007 03:29 PM UTC:

I will match the brutal simplicity of a Rook and a Commoner against the elegant tactics of a pair of Nightriders. One problem with the value assigned on this page to the Nightrider is: 'Repeat Infinite' does not do much for a piece that will bump into the side of an 8x8 board after 2 or 3 leaps. But on a 10x10 board a pair of Nightriders gets stronger, probably equal to at least a pair of Rooks.

Why do I calculate twice as many points for the Commoner and the Silver General as this page does? Following Ralph Betza, I give a higher value to diagonal moves than to orthogonal moves - and then assign huge penalties for being colorbound. Note that the mostly orthogonal Gold General has six moves, which makes it a little stronger than the mostly diagonal Silver General (five moves). Also Betza values the ability to move one square highly. The Queen is certainly worth less than three times as much as the Commoner.


Mark Guo wrote on Wed, Oct 21, 2009 09:39 PM UTC:
I don't know where is the page where you can create pieces. Can you help me? Thanks

Matthew wrote on Sat, Dec 5, 2015 10:19 PM UTC:Average ★★★
I appreciate the effort you put into this, but it's incredibly broken and
messy. Allow me to demonstrate: I create a piece called "Rifleman", since
it is like a more advanced version of the lance - Orthogonal forward (3),
repeat infinite (*5), dual strike move (*8), rifle capture (*3), doesn't
move (/2), single chance move (/4), kamikaze (/4) =10+3*5*8*3/2/4/4=21.25.
The FIDE army costs 110+150 (remember, the king has a *5 mod for being
royal)+2*(70+42+45)+8*15=694 points. 1 rifleman kills the king immediately
(start on e8, capture pawn on e2, capture king on e1). For total,
guaranteed wipeout of the FIDE forces 16 riflemen are required, costing 340
points total, not counting my royal piece, which might as well be immobile
(and that's overkill, I'm sure I could optimize the cost further).
I suggest making these changes: Make rifle capture/igui VERY costly (*5?);
lower kamikaze and single chance reduction to points (/3 each?); explain
powers more clearly, I didn't fully understand the system (WTF is poison
etc.). From what I do understand, they have great potential for abuse too
(a rank of passive burn lances...); metamorph means exactly what?; remove
INVULNERABILITY/can't be captured (for obvious reasons), or at least allow
it only with pieces which can't capture (like immobilizer, ghost) and
aren't royal; define promotes, reproduces, spins and spawner more clearly.
Oh, checkmate (almost) equals can't be captured, have you ever tried
checkmating a mere bishop? I'd suggest removing it, as it's essentially a
duplicate. Poisoned Shogi pawn rush might also be a tactic, so please nerf
it somehow (maybe only working on pieces of equal or lower value than the
poisoned piece?) Ah, so here is how the poison aura works. Strangely
enough, the poison aura is much cheaper than the poisoned modifier (*2 vs
*3, applies only to a part of the piece's cost). The most broken strategy
I can think of is using ranks of immobile pieces with a - 10 pts cost (1/4
bound, can't capture) with an ally-repelling poison aura king to launch
them into the enemy camp. A fix for lot of the broken builds I listed would
be introducing obstacles on the game board (so that you actually have to
purchase sideways moves). Despite all my critique, this is a great system,
and I can see you've put a lot of work into it. I think it certainly would
be more fun than e.g. Chu Shogi, although IMO it still needs some
development.

Matthew edit wrote on Sun, Dec 6, 2015 12:47 PM UTC:
Actually, I made a mistake when creating the "Rifleman" piece. It
doesn't need rifle capture (because Kamikaze), and therefore it costs
13.75 pts, which is an even more ridiculous cost. 16 riflemen therefore
cost only 220 pts, which is like 2 queens.
I also suggest the removal of "repeat infinite" modifier; it can be
superseded by "repeat x" and its removal makes lances less overpowered.
However, to balance it, I'd suggest making the cost for "repeat x" less
for further repeats, for example: repeat x, cost *(x for every repeat up to
the 4th, inclusive. For every repeat beyond the 4th, add only 0.5 to the
multiplier). That prevents standard pieces from having a too prohibitive
cost (R7 would cost 76 pts in this system), while nerfing lance spam,
especially on large boards (where all of the long-range pieces need to have
a bigger value anyways so that they don't overwhelm the short-range
pieces).

Barry wrote on Sun, Dec 6, 2015 08:23 PM UTC:
Fixed points for variant pieces makes little sense. There is no possibility getting relative values correct. This system is good reference for making pieces, but only way to get fair game is auction method. Make enough pieces, do not assign values, give each player 1000 credits, randomly put pieces up for auction one after other. Bidding strategy becomes part of game.

name apelle wrote on Sun, May 7, 2023 05:09 PM UTC:

What if a piece move can only capure pieces of the same type, but the rest of moves can capture any piece?

I'm thinkin of a knight that can move like an alfil only to capture another knight


Bob Greenwade wrote on Tue, Jun 20, 2023 06:16 PM UTC:

I'm trying to figure out if trading places is represented anywhere on this list (for example, a Wizard that can trade places with any friendly piece sitting on a square that it can move to).


13 comments displayed

Earlier Reverse Order Later

Permalink to the exact comments currently displayed.