[ List Earliest Comments Only For Pages | Games | Rated Pages | Rated Games | Subjects of Discussion ]
Comments by jlennert
First expansion is out. 10 new card types, 9 new piece types (including a new Sovereign).
Rules of Chess: Check, Mate, and Stalemate. Answers to frequently asked questions.[All Comments] [Add Comment or Rating]The sort of situation you describe generally results in a draw by one of two rules: The first is the threefold repetition rule, which applies when the exact same game position is repeated three times. For example, if you and your opponent are each moving back and forth between the same two spaces, once you have come back to your starting position after two full loops, the game can be declared a draw. The second is the 50-move rule, which applies when there have been no captures or pawn advances (irreversible moves) for at least 50 consecutive moves of white and black. This is invoked mostly in endgames where the board is very open and so it can take a very long time for (and be difficult to notice when) an exact position is repeated three times. Of course, the game can end in a draw immediately if both players agree, which may cut these conditions short if it is obvious (for example) that the game is going to end in a perpetual check.
Man and Beast 02: Shield Bearers. Systematic naming of divergent coprime radial pieces.[All Comments] [Add Comment or Rating]You write, '[Divergent] pieces are stronger than either non-divergent piece, but weaker than the unrestricted compound piece.' This seems exceedingly unlikely. You seem to be claiming that a piece that moves as a Queen but captures as a Knight is stronger than either; in fact, read strictly, I believe you are claiming that a piece that moves as a Queen but captures as a Bishop is stronger than a Queen, even though it has strictly fewer moves. I believe conventional wisdom is that such pieces (which George Jeliss calls 'snipers') usually have a value that is somewhere between the strengths of the non-divergent pieces, closer to the capturing component than the non-capturing one. (Though I'm certain it is possible to craft examples that violate this rule.)
I remain skeptical even of your revised claim. Even when the components are non-overlapping and 'similar' (though I can only guess what that means), I see no obvious reason that having divergent capturing and non-capturing moves is better than having the same in the general case; only the capture or non-capture will be legal in any given position, so there is no loss due to 'overlap'. In fact, I've seen the opposite argued, on the grounds that a divergent piece is easier to trap, since it is unable to attack enemies in its way. When the combination lifts a special disadvantage (such as colorboundness), that is a special case; though it would need to eliminate a disadvantage from EACH of the components in order for that to make it stronger than BOTH, in general. (And I am unconvinced that lack of triangulation is a disadvantage of any measurable significance.) If the components were instead Crab and Barc, or Rook and Nightrider, what then? Comparing a Pawn to a Point is like comparing a Queen/Knight divergent piece to a Knight; that is obviously the weaker parent (Ferz is already stronger than Wazir, and loses much less from the forward-only restriction). I strongly suspect that the Pawn is weaker than a forward-only Ferz, despite the Ferz being colorbound, because the fF has more possible moves and does not need to capture to change files (the latter advantage being especially important if promotion is allowed). What is your evidence or reasoning for valuing divergent pieces more highly?
Thanks once again for your very interesting computer tests, Mr. Muller! Notice that if this were simply a matter of the K capturing moves being unusually strong and the K non-capturing moves being unusually weak, we should expect mKcN to be weaker than K, but (if I am reading you correctly) the tests say it is stronger. So if you are correct that K's moves have a higher 'ideal value' but are suppressed by K's overall low speed (which would have been my first guess also), then we must conclude that a piece's 'speed' depends significantly upon its capturing moves, not only its non-capturing moves. This seems surprising to me, because I would imagine that capturing capabilities are only exercised in a small percentage of the piece's movements. Alternately, this could be a matter of special synergy rather than a global bonus or penalty--that is, perhaps the combination of speed and concentrated attacks is particularly valuable, but neither component has significant value by itself. Thus, the mKcN, lacking both, is not significantly worse off than either N or K, which only have one each, but the mNcK stands tall with both. Curious.
Rules of Chess FAQ. Frequently asked chess questions.[All Comments] [Add Comment or Rating]Um. They're called 'variants' because they all have different rules. There's no one rule that all variants follow, on that or any other point.
The FIDE rules cover situations that can occur in FIDE Chess. If a variant allows new situations to arise (such as having multiple Kings, or two pieces in the same square, or Pawns on the first rank), it has to specify how to resolve them; they are not covered simply by saying 'FIDE rules apply'. (Ralph Betza once attempted to codify a few rules that were used often in variants as 'Rule Zero', but while he undoubtedly used those rules a lot, I'm not sure whether they're any more common in general than other options, and in any case are not extensive enough to save very much repeating.)
If you're trying to discern the 'spirit' of the rule, I believe it came about something like this:
- Initially, the goal of the game was to capture the enemy King, and 'check' didn't matter.
People got annoyed when an interesting game ended prematurely because one player made a dumb mistake that allowed his King to immediately be captured, so they decided to prevent that by making it illegal. Thus, if you make a move that would cause you to lose on the VERY next turn, you must take it back and do something else (if you have any other choice).
To get the current FIDE rules, you need to add the additional rule that a player who is not in check but who cannot move without placing himself in check (that is, a player in 'stalemate') receives a draw instead of a loss. It's not obvious (at least to me) why this should be so, and historically various players have resolved stalemates in just about every different way you could imagine, but the modern accepted resolution is a draw.
So the 'spirit' of the rule (in my opinion) is 'the REAL goal of the game is to capture the enemy king, but as a safety net, you're not allowed to make any move that would allow your opponent to win on the very next move.' In a variant that nullifies this safety net and allows you to place yourself in check, the most natural rule would be that the game is won by capturing the King, and placing yourself in 'check' is generally a poor strategy but otherwise has no special significance.
Man and Beast 02: Shield Bearers. Systematic naming of divergent coprime radial pieces.[All Comments] [Add Comment or Rating]Good point. Perhaps the value-tests for weak pieces should be performed with armies that include even weaker pieces, such as Wazir, Ferz, or Crab, to reduce the impact of the leveling effect. (One might argue that Pawns should suffice, and perhaps they do, but they are very weird pieces.)
Dragonchess (R). Commercial large chess variant. (16x10, Cells: 124) [All Comments] [Add Comment or Rating]I'm actually quite curious: exactly what rights do you claim that you have to that piece? I am not a lawyer, but my understanding is that game mechanics are not protected by copyright (at least in the US). Furthermore, 'like a Queen but 3 spaces at most' is a trivial and obvious variation on a well-known piece; it has probably been independently invented thousands of times, all over the world, long before Dragonchess showed up. You're also not the first people to give the name 'dragon' to a chess piece, though you might be the first ones with that exact name/movement combination. So what precisely is it that you claim is protected, and what rights do you assert over it?
Concise Guide to Chess Variants. Missing description[All Comments] [Add Comment or Rating]I've never really liked the name 'Prince' for Knight+Ferz. It's supposedly a 'short-range Princess' by analogy to Queen and King, but by that analogy it seems like Princess should be Bishop + KnightRIDER and Prince should be royal.
Dullahan (another of the aos si that predicts deaths)
Rules of Chess: Castling FAQ. Frequent asked questions about castling.[All Comments] [Add Comment or Rating]I've always imagined it is because castling is intended as a development move for breaking board symmetry, not as an escape move to get out of a serious attack. If your opponent has already launched an attack on the King's current position, or if (say) he is using a Rook to cut you off from that side of the board, it's considered unfair to get away or to cross the line of control using a special move. Kind of like how en passant capture was added because the pawn's double-step was intended as a development move to speed up the game, and people didn't like that it was being used to leap past enemy pawns without giving them a chance to intercept, thus altering the game's strategy. But I could be wrong.
Dragonchess (R). Commercial large chess variant. (16x10, Cells: 124) [All Comments] [Add Comment or Rating]I'm no expert on these things, but while the patent's SUMMARY suggests it protects the game as a whole and also separately protects the board and the Q3 piece, the CLAIMS section in the patent never lists the board or the piece as a separate claim--only several versions of the full game. (And the game description appears to cover, roughly, any chess variant played on a cross-shaped board and including the orthodox pieces plus the Q3--or maybe some different specific piece, since 'three spaces in any direction' certainly seems like it could be construed as covering more than 8 directions). So if I understand this correctly, neither the name nor the movement pattern of the piece is protected. (Which makes sense, on account of all the prior art we've already cited.) Though of course any original artwork you made to depict the piece is protected (no surprise there).
Drunken Nights. A toned down version of the Nutty Knights for Chess with different Armies. (8x8, Cells: 64) [All Comments] [Add Comment or Rating]The description of the Colonel on this page conflicts with the description on the Chess with Different Armies page, which permits the Colonel to move as a King in all directions (not just backwards like a Charging Rook).
Suffix Index to Man and Beast. Alphabetic list of suffixes used in the Man and Beast series.[All Comments] [Add Comment or Rating]Grasshopper and Contragrasshopper are certainly similar, and have the same Betza mobility (3.2 with magic number = .7, which would suggest a value around 1.5 pawns). Since both will lose value rapidly as the board empties, I suspect their practical worth will usually just be a question of what you can trade them for. The Grasshopper (but probably not the Contragrasshopper) is very good at threatening undeveloped enemy pieces behind a pawn wall; replacing ANY of white's backrow pieces with a Grasshopper in an otherwise FIDE game allows a fork on turn 1. But without such a favorable opening (for example, if you start the pawns on ranks 3 and 6), I'm skeptical whether they are even as good as a Wazir or Ferz.
Looking only at the best-case mobility of a piece can be hugely misleading, especially when that best case is highly dependent on the exact positions of other pieces. My piece threatens every square on the board, instantly and automatically winning the game...provided that I have 8 pawns all positioned on my seventh rank. Clearly it is stronger than your entire army! You suggest that CG is very good behind a pawn--but as soon as that pawn moves one space forward, the CG is stranded with no moves at all! It cannot even follow the pawn up to regain its lost mobility. A G behind a pawn can only move to the space directly in front of the pawn, but at least it retains that ability as the pawn advances, and can threaten any piece that tries to stand in the pawn's way. You talk about imagining substituting a G or CG for a FIDE piece as if it were some great revelation, but despite your protestations, merely imagining this scenario for a few moments does not magically allow you to accurately determine a piece's value. At least, it does not allow ME to do so, and clearly it does not allow you to produce predictions that are consistent with any accepted method of estimation, nor am I aware of any instances so far in which your unorthodox predictions have been proven true by empirical study. The only actual evidence you have presented so far is an analysis of best-case mobility--which, at 19 squares, should (if it were meaningful) put its value at 1.35 Rooks, higher than either of the two different values you have already guessed! As far as I can tell, your numbers are based on nothing but intuition, and your intuition isn't even consistent. As for me, I'd happily take the Camel in preference to either of these pieces in most positions. By the way, George, your posts would be much more readable if you used an occasional paragraph break.
As has been repeatedly pointed out, CGs directly behind a pawn wall is a uniquely favorable starting position; those CGs have very high POSITIONAL value, but even if they win (which I'm not confident they would) that proves nothing about their MATERIAL value. Start those pawns on rank 3 instead of rank 2 and which side do you think will win? And I can't actually perform your test, since you've removed both kings from the board and haven't stated a new win condition. Which makes me pretty confident that you haven't actually tried this test either, since that issue would be pretty hard to miss if you actually sat down and played. So even if this were a fair test (which it's not), it's not a test that has actually been performed, so to say that the CG is 'measurably' above Camel is just a bald-faced lie. You have yet to measure anything. How about this? Give each side 8 pawns, a king, and either 6 camels or 6 contragrasshoppers. Put the kings on their usual starting squares, and distribute the rest of each side's pieces RANDOMLY throughout their half of the board (I guess we'll forbid pawns on the first rank just so we don't have to pick a special initial-move rule). I won't even require that the camels are split between different colors. Pawns promote to queen. If a king starts in check, his side gets the first move, otherwise pick randomly. I'm sure each side will win some possible random positions. I'm reasonably confident the camels would win the clear majority--but I haven't tried it, so I could be wrong! (I suppose I really should look into getting some software to run computer tests like Muller has been doing...)
Chess 2. Different armies, a new winning condition, and duels. (8x8, Cells: 64) [All Comments] [Add Comment or Rating]The rulebook is poorly written; information is not presented in a logical order, it is difficult to tell in some cases what is a new rule and what is intended merely to explain the implications of another rule (which might not appear until later in the rulebook), and it glosses over several crucial details. Some of the diagrams depicting unit movement are exceedingly confusing (I defy anyone to look at the diagram for Jungle Queen and guess how it moves without reading the text), and one depicts a board position that is (as far as I can tell) completely impossible. Very low marks for editing. The armies are rather interesting; I was expecting pieces similar to FIDE but with different movement patterns, along the lines of Chess with Different Armies, but most of the new pieces in this game have unique special rules. There's only about 10 new things between all 5 armies, though, so if you were expecting CwDA amounts of new pieces per army, you will be disappointed. I don't know whether any of it is balanced; it seems unlikely unless quite a large amount of testing was done (but maybe it was, I don't know). And I wouldn't be at all surprised if a computer could discover a short forced win from some opening positions (even with the simultaneous-action dueling mechanic). In particular, Reaper vs. Reaper looks like just a race for the midline with no significant risk of checkmate at any stage of the game. It is also my suspicion that a computer could play this game very easily. 'Dueling' increases the branching factor (as do some of the armies), but I expect games would be much shorter, and I think computers would be better at handling most of the new elements (especially deciding when it's time for the king to make a break for the midline).
Yeah, I don't necessarily agree with Hubert's criticism, but in my opinion Jepps' behavior in this thread has been indefensible. If you're not prepared to tolerate criticism of your work, you shouldn't make it public. The suggestion that only positive commentary should occur is childish and transparently self-serving, especially for someone who drummed up attention by rating his own game 'excellent' four separate times.
The 'Poor' rating is there for a reason, and anyone who thinks you can distinguish good amateur content purely by the quantity of attention garnered is sorely lacking in experience of such matters.
As far as I can tell, George gets all of his piece stats from about 95% sheer intuition and 5% wildly unrepresentative examples. If you actually went through his numbers rigorously and systematically, I suspect you'd discover that many of them are total fabrications.
He even gives a mate # of 2 for Water Rook, a piece that George himself defined as specifically only moving on light-colored squares (the piece that moves in exactly the same pattern on dark squares has a different name), which therefore obviously cannot force a mate with any number! That's like giving a mate # for Bishops that are all on the same color.
...and even if he hadn't defined them in this bizarre way, two is definitely not enough of them to force a mate in general. I believe two Dababba-riders can sometimes force mate, if they are on different colors and can cut the enemy king off from the 'safe' edge of the board, but the Water/Land Rooks are lame Dababba-riders, and therefore cannot control an entire rank/file even when working together.
Incidentally, if we're playing fox-and-geese with rotating spearmen, I think you'll find that 3 spearmen + king is sufficient to win most reasonable starting positions (use two spearmen pointing forward to confine the enemy king to a corridor and then just chase him down).
'Lame' is Betza's technical term for a piece that can be blocked on a square that it cannot land on. Say you've got a DD on a6 and another on h6. What do you do when the enemy king moves to b7? If they're full Dababba-riders, the one on a6 just moves to g6, and you still cover the whole rank. But if they block each other, you're stuck. If he moves anywhere on the same rank, he blocks the attack from h6 to b6, and the king slips through. If he moves anywhere on the same file, he is no longer covering c6, and the king slips through. If he doesn't move, the king captures him.
Black's goal is to draw. He doesn't have to avoid threefold repetition; white does.
If you suppose your king will always maintain opposition to the enemy king, then it can block the rank with no help at all. But if you don't maintain opposition, then the enemy king necessarily has a head start towards one side or the other, and can break your line there before your king can get close enough to defend, since you are vulnerable on both sides. And since maintaining opposition at all times will require all of your moves, it cannot be part of any plan to force mate. But a pair of DD (with leaping power) plus a king CAN force the enemy king back using zugzwang, unless I am much mistaken. For example, suppose white DD on a2 and b2, king on c3, black king f3. 1. Kd3 Kg3 (black tries to stay on rank 3) 2. Ke3 Kh3 3. Kf3 Kh4 (black is forced back a rank; only legal move) 4. Kg2 Kg4 5. DDb4 Kg5 (King threatens f3, g3, and h3, DD threatens h4 and f4) 6. DDa4 For thoroughness, we should also advance our own king to rank 5 without breaking the line, so we can repeat the process... 6. ... Kf5 7. Kg3 Kg5 (black tries to maintain opposition) 8. DDc4 Kf5 (white loses a tempo to break opposition) 9. Kh4 Kg5 10. Kh5 The final mate is only slightly more complex. Suppose white DD a6 & b6, white king c7, black king f7. 1. Kd7 Kg7 2. Ke7 Kh7 3. Kf7 Kh8 4. Kg6 Kg8 5. DDf6 Kh8 (f8 is threatened; black confined to corner) 6. DDa4 Kh7 (white loses a tempo) 7. DDa8+ Kh8 8. DDf8# (or DDh6#) I trust it's easy to see why this doesn't work if the DDs block each other.
If you're calculating mate #s based on your own variant chess laws instead of the orthodox ones, you should have stated that up front. And if your variant laws include removing the 50-move draw and making repetition illegal instead of a draw, then the numbers you are calculating are likely much closer to helpmate numbers than forced mate numbers. Eventually, ALL non-mate positions will be exhausted, and black will be forced to collude in his own checkmate by process of elimination.
25 comments displayed
Permalink to the exact comments currently displayed.
There's an entire sample game available from the publisher's site, including commentary and illustrations, if you're interested.
I could see some value in adding brief examples to this page, but doing so seems awkward, because the editing form doesn't allow me to create new sections, just edit the contents of the existing ones.