The site has moved to a new server, and there are now some issues to fix. Please report anything needing fixing with a comment to the homepage.

The Chess Variant Pages

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

Comments/Ratings for a Single Item

Later Reverse Order Earlier
Gnus. Makes (1-2)-jump or (1-3)-jump.[All Comments] [Add Comment or Rating]
Charles Gilman wrote on 2011-09-18 UTC
I just found a comment that I had meant to post sooner:
	I never meant that being weak was per se a reason why a piece might be neglected, I was thinking more in terms of weakness combined with complexity. The Antelope is indeed in the Piececlopedia but the Fiveleaper, an Antelope compound with another colourswitching piece, is not - making it slightly puzzling that the Fiveleaper's colourbound dual the Root-50 leaper is.

George Duke wrote on 2011-07-21 UTC
So Zillions estimates for 10x10 Piece Promotion Games are Gnu 5.94, Gazelle 5.54, Bison 5.20 versus the other basically comparable estimate here Bison 6.2, Gnu 6.1, Gazelle 6.0 re-ordered for reasons stated. Camel leg and Zebra leg differentiate fast; on 8x8 start a Camel a1 to b4 to e5 to f8, three moves of Gnu(CN) staying on the board and unable to get closer than two away from the opposite corner h8. On 8x8 start a Zebra a1 to c4 to f6 to (h9 or i8 both nonexistent) and he is outside the board beyond the corner on the same third move, like three long-direction moves of Gazelle(ZN) or Bison/Falcon(CZ). The latter two need bigger board or another board to make just three crosswise Zebra moves without doubling back. Gnu as compound of duals, Knight and Camel, triangulates for what that is worth. Triangulating is an attribute, even a happenstance, it is not attractive to give most piece-types. See_Gilman's_catalogued_oblique_leapers, where in the series and his own cvs Gilman certainly favours bi-compounded duals. However, Falcon/first-use-of-Bison and Gazelle themselves are based on other concept of organization than duals, as would Overby's Beastmaster leaping piece-types diverge from that particular standard enforcing triangulation. The usage this thread is correct even in OrthoChess vocabulary where only King and Queen triangulate, return in three. As a tactic there the meaning has difference in extension: Tactic. The ''Ungulates'' paragraph starting ''As well as the Gnu...'' mentions that Bison/(Falcon) returns in five. --The other three fundamental chessic units, those Bishop, Knight, Rook, return in four.

Ben Reiniger wrote on 2011-07-21 UTC
I would probably agree that forking isn't increased with triangulation; I hadn't though about it long.  On the other hand, if the guarded square is friendly, triangulating means attacking some new squares while still guarding the old one.  I don't think in general that attacking a new square or attacking an old square is more advantageous, but having _both_ options is certainly nice.

Anyway, I hadn't really looked at which pieces were being discussed below yet.  You're correct that the bison doesn't triangulate (despite not being colorswitching), so my argument doesn't help in that discussion.  I wanted in my previous post just to present one reason why colorswitching may be detrimental.  (As a side note, the interaction of colorswitching with colorbound pieces is interesting.)

I agree about the bison vs. centaur.  The bison is an annoyance at range, and the centaur is strong locally.  In For the Crown, I have found the bison to be particularly nice in the early to midgame, especially since the attack cannot be blocked by dropping a piece adjacent to the King (in particular with the Guard's order).

Jeremy Lennert wrote on 2011-07-21 UTC
Unless I am mistaken, the Bison cannot triangulate either.

Attacking the same square after a move is an advantage, but attacking a different square after a move is also an advantage.  I'm not currently persuaded that the former is better.  You mention forking power, but attacking a new square seems likely to be better for that--if one of the pieces you 'fork' was already threatened and your opponent chose to leave it in place (and you chose not to capture it), then the fork isn't likely to distress him overmuch, is it?

It is interesting to compare the Bison to the Centaur (WFN, knight+king).  Both have 16 moves, but the Centaur's power is concentrated, while the Bison's is dispersed.  If you drop them on a random location on a crowded chessboard, the Centaur reaches more squares on average (because its moves are less likely to be limited by the edge of the board), but the Bison has more squares it can reach within 2 moves.

I've played with both in For the Crown, and found the Centaur effective for defense and (with support) for forcing a checkmate, but the Bison appears to have far more forking power and makes an excellent harassing piece (though part of this advantage seems to come from 'stealth', having more moves that are not shared by enemy pieces).  Though perhaps players with greater skill would draw different conclusions.

Ben Reiniger wrote on 2011-07-21 UTC
One disadvantage of color-switching pieces is that they cannot triangulate; other pieces also lack this ability, but colorbound ones necessarily lack it. (I'm not sure if what I think triangulating means is common usage. I mean that a piece can move to its original position in three moves. The point is that the piece can be guarding a given square, move once, and still be attacking the square. This allows the loss of tempo and perhaps also increases the possibility of forking.)

Christine Bagley-Jones wrote on 2011-07-20 UTC
AndR, i have no idea how zillions comes up with these numbers, it gives a pawn in this game the rating of 2,752.
It is free to become a member of this site, and your comments will be posted straight away instead of the wait, you should consider this :)

George Duke wrote on 2011-07-19 UTC
Right, they are three very close. Orthodox chessists do not dwell on relatively trifling 0.05, 0.10, 0.20 differences say in FRC line-ups between Knight and Bishop or minor rules changes affecting stalemate or whatever, do they? Good moves are more important the players, and for designers good piece-types and rules that jibe and reinforce, not exact valuation. (For instance, Winther's dozens of bi-furcators he reports as ''about 3.0'' or ''estimated at 4.0,'' close enough for initial purpose.) Yet specifically here, the disadvantage of forced colourswitching is in view of other pieces' like Bishops being colourbound. Gazelle cannot capitalize on positional advantage against such colourbounders generally speaking. Bison and Gnu can elect to switch or not, and Gazelle cannot.  It takes play of Bison to internalize the power of long-range forks both Gnu and Gazelle have not the same ability. 'Bison > Gnu > Gazelle' is pretty clear on our intermediate boards, meaning over 72, notwithstanding Zillions' formula. Only three steps of Bison, either Camel or Zebra, lose nothing in versatility.  More so striking would be Bison over the others on greater than 100 squares. Anyone playing Omega Chess on 100(104) for example knows how lost get both Wizard, whose reach is 3 only half the arrival squares, and Champion to the line pieces. However, the over-100-squares set should result in Gazelle eventually overtaking Gnu for that very range-advantage of Zebra ''farther out'' than Camel. Over-all, we are talking most board sizes/piece mixes within best common sense, 80-120 squares, of values like 6.2/6.1/6.0 the three bi-compounds near equal. [On a Jupiter 16x16 or even possibly a Typhoon 12x12 start about Bison 7, Gazelle 6, Gnu 5 and see what happens.]

Christine Bagley-Jones wrote on 2011-07-19 UTC
As George said, they are all very close. I do like Jeremy's comment too, 
'The long move gives you speed, but the shorter gives you more

Jeremy Lennert wrote on 2011-07-19 UTC
What makes you think that a color-switching piece would be weaker than a non-color-switching one?

It is easy to understand that color-boundness is disadvantageous because it means there are squares you cannot reach even with an unlimited number of moves, but color-switching implies no similar disadvantage that I can see.  A knight can famously tour the entire board.

If I were to guess, I would say the Bison is likely weakest of the three, because I conjecture it is more useful to have both a shorter and a longer move than to have two long moves.  The long move gives you speed, but the shorter gives you more maneuverability.  But this is only a guess.

George Duke wrote on 2011-07-19 UTC
Bison > Gnu > Gazelle should be right order of three very close bi-compounds on board greater than about 72. They should each rate nearest 6.0 than any other whole number? Then too I still say Falcon(reducing to 5.0 after so many moves) > Rook(5.0) very slightly on 72-100, and Muller has temporarily ''proved'' that wrong; so Zillions too may have some different rationale hard immediately to dispute. Besides colourswitching handicap of Gazelle, Gilman would say correctly that longer-range weaker value does not kick in yet at Bison (1,3 + 2,3) distance. Low piece-value Antelope(3,4) is under 'A' at Piececlopedia, so Gilman knows that one remark of his not up to his usual excellence, and that clever subtle weakening, combined with binding pairwise, in fact makes the best p-ts for Piececlopedia or Short-Range projects, not the implied other way around.

Christine Bagley-Jones wrote on 2011-07-19 UTC
I dont really understand what you mean when you say '.. the Gazelle would be weaker, and that that was why it was not popular enough to merit a Piececlopedia page.'
I don't think a piece not mentioned in piececlopedia means something negative about it. 
I think Fergus is in charge of this page, but anyone can write about a piece and put it up for being added to the page. There are many fairy pieces in existence and if one person is in charge of the page, i can easily see why it doesn't have every different piece under the sun added, too big a job if you have a life hehe. 
The ratings i gave is what 'zillion's' gives the piece, i am not saying that this is correct either, it just give's you an idea, i'm sure there are people who probably can give more 'correct' answers to the strength of these pieces.

AndR wrote on 2011-07-19 UTC
Christine, thanks for responding so quickly.  I do not currently have
Zillions, but I may check out your game recommendation if/when I get it. 
From Zillions, you give the ratings: Gnu 16,338; Gazelle 15,245; Bison

Is this in terms of 'pawn value = 1'?  In that case, I would assume your
commas are decimal points and that we're talking about a Gnu being worth a
little over 16 pawns, a Gazelle around 15 and a Bison around 14.  That
seems much stronger than I was expecting!

But I wonder if I misunderstand how Zillions rates pieces.  Is there some
other unit in use here?

Thanks in advance for the clarification.

Charles Gilman wrote on 2011-07-19 UTC
I am surprised to see a higher strength rating for the Gazelle than the Bison. I would have thought that as a compound of two colourswitching pieces, and therefore colourswitching itself, the Gazelle would be weaker, and that that was why it was not popular enough to merit a Piececlopedia page. If it is really not that weak, is anyone thinking of adding a page any time soon? My latest submission, Index G to Man and Beast, describes the name Gazelle as an 'established name but for some reason lacks Piececlopedia page'

Christine Bagley-Jones wrote on 2011-07-19 UTC
The Gazelle is a pretty standard name for the knight/zebra compound.
There are many pieces not mentioned in Piececlopedia.
If you have Zillions of Games, download my game 'Piece Promotion Games 2', it has the Gnu, Bison and Gazelle in it's variants. Zillions rates these pieces as follows ..
Gnu 16,338
Gazelle 15,245
Bison 14,316.
Also you can see the Buffalo (knight/camel/zebra) and Squirrel (knight/dabbaba/alfil) in this game.
Another place to look at different fairy pieces is

AndR wrote on 2011-07-18 UTC
I've been trying out the Gnu as a chess piece and really enjoy playing
with it.  One variant it does well in replaces knights with Gnus and
bishops with FADs and is otherwise orthodox.  A pretty simple change but a
very different game than FIDE Chess.  The Gnu seems to be a pretty strong
piece, with 16 squares it can move to from the middle of the board.  I
wonder if somebody with some experience estimating piece values could say
what they think it would be worth, relative to the orthodox pieces.  If a
pawn is worth one point, how many points is a Gnu worth?

I have that same question concerning the other oblique leaper compounds. 
In particular:

* The Bison (Camel+Zebra)
* The Gazelle (Knight+Zebra)

(Incidentally, I've noticed that the latter piece isn't in the
Piececlopedia, and I wonder if 'Gazelle' is even an agreed-upon standard
name for it or where it comes from, etc.)

I suspect the Gnu is the strongest of those 3 pieces, and I don't know
whether Bison or Gazelle would be next-strongest.  The Gazelle has to
change color every time it moves, which is not the case for the other two
pieces, which might put it at a disadvantage.  But then again, both of the
Bison's attacks are longer than the knight move, which makes me think it
might be a weaker piece.  Has anyone considered these questions?

Thanks from a new fairy chess amateur.

Charles Gilman wrote on 2003-07-05 UTC
While looking through the large variants I have found three more using this piece. The list of those I know now stands at: Gnu in Mark Hedden's Ganymede Chess and Io Chess, Wildebeest in Wayne Schmittberger's Wildebeest Chess and in Glenn Overby's Abecedarian Big Chess and Promo Chess, and Cavalier in Stephen Sava's New Chess. There may well be others.

Charles Gilman wrote on 2003-05-17 UTC
My previous comment needs clarification. When I said that the Gnu 'can lose the move, unlike its components', I had a square 2d board in mind. On a 3d board the Camel itself can lose the move, though no faster than the Bison. The Knight cannot even lose the move in 3d, nor can any root-odd elemental leaper, but root-even ones can if their leap does not pass through the centre of a cell of the other Bishop colour, e.g. Ferz and Alfil but not Dabbaba. Note that Unicorn colourbinding is irrelevant as a cell on a 3d board is orthogonally adjacent to cells of all three other Unicorn colours.

Glenn Overby II wrote on 2003-05-04 UTC
On the naming of combined leapers: I developed a family of combined leapers
for Beastmaster Chess.  Positing the Horse's move as three squares
including a turn (another way of describing a 1,2 leap), I used pieces
with four, five, and six square moves including a turn.

The colorbound Roc is (1,3) + (2,2), Camel+Elephant.
The colorchanging Pegasus is (1,4) + (2,3), Giraffe+Zebra.
The colorbound Wyvern is (1,5)+(2,4)+(3,3).

None of these names is in =common= use elsewhere.  Pegasus found its way
into ximeracak. as well.

Charles Gilman wrote on 2003-05-04 UTC
Further to my previous remarks, this piece is called a Gnu in Mark Hedden's Ganymede Chess and Io Chess. Unlike its components it can lose the move - it can return to a square in 3 moves or move to an adjacent one in 2. It is a stronger move-loser than the Bison, which takes 5 moves to return. Is it true that the Knight-Zebra combination (which cannot lose the move because it always changes square colour) is called the Gazelle, and are there names for any other combined leapers?

Charles Gilman wrote on 2003-03-30 UTC
Michael Howe is right, the singular is Gnu and Wildebeest is an exact synonym. In Wildebeest Chess it is part of a matching for the ancient Knight what recent standard pieces are to the equally ancient Rook. The Camel move is simply a Knight move turned through 45° and multiplied by root 2, as a Bishop move is to a corresponding Rook move (consider the triangle b2-a3-c4). Thus this piece combines the Knight and its colourbound dual as the Queen does the Rook and its.

20 comments displayed

Later Reverse Order Earlier

Permalink to the exact comments currently displayed.