Comments/Ratings for a Single Item

Inspired by Dragon Chess, no? This game is pleasant to see. The difficulty is probably to well balance this game. Have you tried some games already against opponents or an AI, or are you simply "painting" boards with nice-looking pieces and imagining moves? Playtesting is important.

For most of the things I've thrown together, I've done some sort of calculations and/or careful comparisons to existing variants in an effort to keep things balanced. This is the only one that's mainly, as you say, "painting" boards with nice-looking pieces, so I'm not at all sure how balanced this is; I fear that I may have overpowered things, most significantly in the Air pieces.

I've actually never seen a copy of the Dragon Chess rules; this was more inspired by noticing a surprising number of pieces that are named for sea and air creatures. Learning of the Ship sealed the idea; coming across a Strato Chess set on Amazon (I had one when I was a kid) spawned the Air level.

I'd like to try this against an AI, but for some reason this old programmer is finding the idea of doing so rather overwhelming, and the sheer number of pieces in this is only a part of the reason. That said, any recommendations that would work for this one would be welcome.

Dragonchess is described in WP: https://en.wikipedia.org/wiki/Dragonchess It also has 3 levels subterrannean/earth/air.

Personally I use ZoG to test my variants, whatever the limitations reported to this package, it is pretty useful to make these tests. Unfortunately it is not so widely available those days.

Interactive Diagram, provided on this website, is a very useful tool too.

I'm going to give a shot to ID for Vanguard Chess, but I'm not sure how to do one for a 3D game like this one.

With as many pieces as I "borrowed" from other games here, I wonder if anyone else would want to "borrow" some of the pieces that I made up myself.

The Sparrow (oQcK) seems like the most likely, though the Jellyfish, Manta, and Seahorse all have their own interesting characteristics.

I could make a Note of it in the text, if I can figure out the (2D) Betza for those other three.

EDIT: I think I figured out the Seahorse as nN, and the Manta as B2caibW. For the Jellyfish, I'm not sure there's a way to write in Immobilizing with XBetza.

I only just realized (d'oh!) that the starting positions allow the Ravens to attack each other immediately, so some rearrangement is called for....

I'm still finding the placement on the Air and Sea levels a bit awkward, with non-Pawn types in the front rows. I'm thinking of either:

1. dropping two main pieces from each (most likely Manta and Seahorse from the Sea, and Raven and Pegasus from the Air); or
2. adding four more Fish to put in front of the Crabs and Jellyfish, and four more Sparrows to place in front of the Geese and Pegasi; or
3. abandoning the 8x8 boards for 12x12, putting all of the non-Pawn pieces on the back row and filling the second row with Fish and Sparrows.

I'm most likely to do the #1. #2 makes for (IMO) an overly crowded board, and #3 is unattractive because I wanted to extend the use for three 8x8 boards as found with Strato Chess/Chess3/etc.

Another argument against #3 is that I'm already tinkering with something along those lines, that would also add two pairs of "split" levels (one 12x6 on each side) for Land (between Waves and Air) and Caves (between Sea and Waves), with even more beasts, military types, mages, magical creatures, and even undead, with 144 pieces per side. On the other hand, that's one I'm mainly building as a mental exercise rather than for actual play (though I'd definitely give it a whirl given the opportunity).

There's even option #4: ditch Aquachess altogether, and just build that huge project.

EDIT: I think I figured out the Seahorse as nN, and the Manta as B2caibW. For the Jellyfish, I'm not sure there's a way to write in Immobilizing with XBetza.

Why do you describe the Seahorse as using two or three steps to make a Knight move? What would be the path of three in going from e4 to d6?

In the XBetza for Manta you can omit the N, as W is a leaper atom, so it always only does a single step. The i is only needed in rifle captures of sliders.

Note that oQ indicates a Queen on a cylinder board, which has K as a subset, so the addition cK is pointless.

Immobilizing is not a move, so there is no XBetza notation for it. In the Interactive Diagram you can specify it by tracking the immobilizing piece, and define spell=freeze. The ID is not really made for 3d variants, though. I suppose you could use the work-around of defining an L-shaped board of three 8x8 sections, separated by sufficiently wide regions of holes that prevent all 'level' moves from crossing. And then define the inter-level moves as very long leaps that must cross them. Some moves in this variant might be too irregular for this, though, and it would require dedicated scripting to extend the spell zone for freezing to neighboring levels.

Seahorse: I probably got overly specific on the text description here (a bad habit in my writing). Basically it's an unobstructed Knight. I'll fix that. And, now that I figure it, only two moves (one diagonal and one orthoganal, in either order) should be needed.

Manta: I didn't have an N in the Manta's XBetza. But anyway... if the i isn't needed, should it then be B2cabW?

Sparrow: My error. It should be mQcK. (I don't know why I used o.)

As for in general, I don't think I'd even try to code level-switches in XBetza; I don't think there's even a way to modify that system for it. I was just trying to show the 2D XBetza from pieces that people might be interested in "borrowing" from this variant, since I did the same with so many (expanding most of them to 3D).

There is no Betza-like system for moves in 3d; even if there was it would probably not be helpful in a variant like this, where layers have different properties, rather than representing an isotropic 3d space.

I was just contemplating how far one could push the Interactive Diagram towards handling a game like this through the trick of mapping all levels to a 2d board. (Something that in practice would have to be done for playing it anyway.) E.g. an 18x18 board could have the waves from a1 to h8, the sea from k1 to r8, and the air from a10 to h18, and the rest inaccessible 'holes'. A vertical move between waves and air would then be a (0,10) leap, or vWXXX in XBetza.

I can make that worse for you:

In the huge 5x(12x12) variant that I mentioned, the Sea, Waves and Air are levels 1, 3, and 5. The new levels are Caves at 2, and Land at 4. They're not parallel to the odd-numbered levels, though; they're split down the middle between Left and Right, with their inner edges in line with the other boards' right edges. And when a piece can move from one of the outer boards to one of the inner or vice versa, it does so as if there was no barrier.

If I ever truly build that one, never mind IG or the Java Applet; I might try to make a Zillions file for it, though.

Even so, if I were to try to make a Betza code for something like Aquachess, I'd make it as an "eligibility" marker on the Pieces (something like the type-sensitivity idea already proposed on the Betza page). Those restricted to one level just wouldn't have vertical moves; those with vertical moves that can only go to two levels (which is everyone but the Dragon) would have a restriction marker against Air or Sea.

(And I think you can probably see now why I said I wouldn't even try with it!)

I do notice that XBetza doesn't use a lot of punctuation marks; perhaps something like +, &, %, or ^ could be utilized to indicate a level change. If that sounds like a possibility, I'll make a fuller proposal on the Betza page.

The problem with Zillions implementations is that not too many people can use those.

About Zillions, I agree. However, being coded for ZoG seems an important feature for Fergus. Btw, I uploaded the zrf for Metamachy and I posted a comment to make sure that I did it right. However, I got no answer.

Well, I did say only that I might make one. And even that much is mainly because neither IG or the Java Applet are well suited for 3D games (made worse by the "tiered" layout of the big one).

As far as 3D (or in general non‐square) Betza extensions/analogues go, the only real problem is the different kinds of direction which don't necessarily correspond to the square‐board ones. But at least for Atoms that shouldn't be too big a problem: there's still plenty of capital letters left to go around; probably even enough to cover outlandish things like hex‐prism geometry if necessary

Directional modifiers (extending f, b, l, r, s, v, and the like) are more of a problem given the paucity of lower‐case letters left over

Yeah, no kidding, Bn Em! That's why I (1) said I wouldn't want to try programming the function in, and (2) suggested using punctuation marks.

Since there's been no further feedback on this, I've taken out the "This is a work in progress" sentence under Notes.

Let's give this a go, OK?

However the Sparrow (mQcK) is Duke from Dan Lee's asymmetric Empire Chess (which is playable on Pychess)

However the Sparrow (mQcK) is Duke from Dan Lee's asymmetric Empire Chess (which is playable on Pychess)

There are so many different takes on what should be the Duke that I'm not even going to bother with that title. If I ever do, it'll probably be the Musketeer version (W2N).

Nonetheless, thanks for the info.