Game: TZAAR

Thread: Re: suggested opening strategy

brdparker — 2008-08-20T03:19:26+00:00

Alan Kwan wrote:
I agree that there are reasons to play double capture moves, and I do play them myself. However, I maintain that one should make such move only if he has a good, clear reason. Some of the ones you stated are good reasons. Though, one rarely has a good reason in the first several moves of the game.

Okay, I'm with you there, one rarely has a good reason to stack early-game, and the double-take reasons should be very specific. Since responding, I've played a couple of games (with your article in mind) and have shifted the proportions of stack-double take from about 70%-30% to about 85%-15% so you've converted me to a point

Alan Kwan wrote:
Stacking near an enemy piece often has as good an effect on threatening the piece and "limiting opponent's options" as capturing it, unless that piece is covered (protected) by an enemy stack. For example, I won't play a double capture move against a center Tott (standard setup) to "limit stacking options", since it's better to just make a stack near it so that he can't stack on it. And if my opponent does that against me, he's just wasting his time as I stak Tzaar on Tzaar, or Tzaar on Tzarra (which is no worse than his capturing a Tzaara instead of a Tott, and then I stack on the Tott).

Touché. I see your point

This game is so great...

Thread: Re: suggested opening strategy

Alan Kwan — 2008-08-20T01:52:30+00:00

I agree that there are reasons to play double capture moves, and I do play them myself. However, I maintain that one should make such move only if he has a good, clear reason. Some of the ones you stated are good reasons. Though, one rarely has a good reason in the first several moves of the game.

Stacking near an enemy piece often has as good an effect on threatening the piece and "limiting opponent's options" as capturing it, unless that piece is covered (protected) by an enemy stack. For example, I won't play a double capture move against a center Tott (standard setup) to "limit stacking options", since it's better to just make a stack near it so that he can't stack on it. And if my opponent does that against me, he's just wasting his time as I stak Tzaar on Tzaar, or Tzaar on Tzarra (which is no worse than his capturing a Tzaara instead of a Tott, and then I stack on the Tott).

Thread: Re: Tzaar in the top 200!

TheBrainDrainer — 2008-08-19T20:34:16+00:00

I have all the Gipf games (and potentials) and currently like Tzaar best, followed by Yinsh.

Thread: Re: Tzaar in the top 200!

trapeze — 2008-08-19T19:15:06+00:00

I don't think so. It's a very good game, but not quite there with Yinsh and Dvonn (and Gipf for that matter).

Thread: Re: Tzaar in the top 200!

Yokiboy — 2008-08-19T18:03:36+00:00

I've only played YINSH so far, but am looking forward to trying all the other GIPF titles.

Thread: Re: Tzaar in the top 200!

russ — 2008-08-19T04:54:19+00:00

For me, it already has.

Thread: Tzaar in the top 200!

BrenoK — 2008-08-18T21:14:23+00:00

Watch out, Agricola!

So, do you guys think Tzaar will top Dvonn? Or maybe even Yinsh?

Thread: Re: suggested opening strategy

brdparker — 2008-08-16T14:59:05+00:00

Alan Kwan wrote:
The same does not apply to the player whose stack is half a piece shorter. If your stack is half a piece shorter, keeping building it single-mindedly will just give you a taller stack which is always still half a piece shorter than the enemy stack (unless you opponent makes a dumb mistake and neglect or fail to keep growing his stack against yours, which will not happen according to the basic game theory assumption).

Interesting terminology, the half-piece advantage. I think I'll use that more often. That;s what I love about these games; Burm set up a set of genius systems that the players not only have to figure out strategies for, but have to develop a glossary based on what they "feel" during a particular move. Eventually it will become standardized, I'm sure, but for now, it comes down to a number of individuals going under trial and error and having a great, brain-burning time in the process

Alan Kwan wrote:
Note that playing a double capture move is still not an option: it is a last resort when your stack has to escape, or a positional move (not a material move) for cutting off the enemy stack's movement (i.e. you should have a clear reason for doing so).

I don't agree here. I often use a double capture for a number of possible reasons:

a) I'm trying to kill off as many of one type as quickly as possible. Now you could say that I could just take one (say, a tzarra) and then strengthen, but take this example: there are 4 white tzarras left on the board (all singletons or doubles). You have a triple or quadruple black (of whatever type) and have noted that at this point in time, at least one of every set of your is secure (that is to say, unless you make a horribly stupid move, you cannot put yourself into check).

Your large stack is within striking range (zone of control, I believe has become the phrase of choice) of 2 of your opponent's tzarra/tzarra stacks. Thus, you take the other two, as it allows you to focus your efforts on the 2 remaining, the ones that endangered by your monster stack. This is obviously taking into account the ZoC of your opponent's large stack(s) and how in danger your own sets are.

b) Kill of tots so as to decrease stacking options.
So obviously at setup, you're trying to kill off your opponent's tzaars. Usually (unless he plays stupidly), there comes a point in the game where you switch and start pegging the tzarras. Thus, if he knows this, he should never stack a tzaar onto a tzaara, because that would make the appeal to switch even greater. So what's left? He can stack onto other tzaars or other tots. By default, you're going around killing tzaars, so that's obvious, but the inner tots (if you're playing basic setup) are prime candidates for the double-kill since it forces where you opponent may stack.

c) Cut down on opponent's mobility.
Assume the setup where you know your opponent wants to move a largish stack into one of the corners (and has a singleton there he can stack on). That singleton is within ZoC of one of your singleton tots. You can either take-strengthen (by taking the singleton and then moving onto one of your own with the same piece, thus nullifying the large stack's ability to move into the corner), or take-take (by taking the singleton and than another singleton is you don't have anywhere you can strengthen)...

Just my thoughts on the matter. I'm curious to hear your response

Other than that though, I liked the article. Good techniques!

Image

wizardless — 2008-08-11T08:46:57+00:00

game of tzaar at the hatepe mid winter gaming retreat 08

Thread: Re: stacking question

russ — 2008-08-04T17:36:42+00:00

THE SECOND MOVE OF A TURN wrote:
b) Making a piece stronger
3. Any of your pieces or stacks can jump on any of your other pieces or stacks.

Thread: Re: stacking question

desmodus — 2008-08-04T14:48:42+00:00

Yes you can use the same piece. The only restriction is to first capture.

Tzaar is definitely my favorite abstract game.

A must have!

Thread: stacking question

joelbear — 2008-08-04T14:15:01+00:00

can you stack the same piece you used for capture in move 1 of your turn?

thanks

Joel

Thread: Re: Opening strategy

Phlegm — 2008-08-03T00:06:24+00:00

I like this game better with the random setup. The suggested setup just seems like there will always be the guy who has it figured out. Kind of like playing a grandmaster in chess. If you don't make the "correct" move then you're screwed.

With the random setup each game plays differently so there is no "best" moves in the opening. It's also one of the things that I like about DVONN as well.

Thread: Our first few games of Tzaar

mjrobertson — 2008-07-21T09:39:04+00:00

My wife and I played our first games of Tzaar yesterday. She was entirely new to it whereas I had previously read the rules and played a solo demo game to get a feel for the mechanics. Rule explanations took about 2 minutes, a minute to set up the board (randomly) and then we were playing.

Game 1 was a tense affair, all concentration and no talking besides the odd grouchy comment when one of us made a sneaky move. After the first few moves my wife had gotten into the swing of things. It really is easy to pick up and play. Mid game progressed with various exchanges, the odd blunder, and the odd clever move.

The end game was tight, both of us having few counters left – I had stronger stacks but my wife had more stacks. A couple of times I was a move away from defeat, but always managed to worm my way out after a good think. There really are so many different possibilities that it's hard to see everything all at once. Finally she had me cornered and I lost from only having 2 piece types left. And I was pretty sure at that point that I was going to win on my next turn!

We immediately set the board up again for a rematch. This game was much the same as before - no talking, just intense concentration. I eventually won by capturing her last Tzaar in the end game.

Then it was on to the deciding game. By now we were both forming tactical ideas and looking more at the overall board situation. I could now see that despite the game's simplicity there are a lot of tactical similarities to chess: opening/closing lines, discovered attacks, forks (to a certain extent), sacrifices, mutual protection of stacks, and even some plan ahead calculation of what might happen in various areas of the board. Board position also seems to play a part, although I couldn't really tell whether it was better to stay near the centre or on the outside.

I could even see vague similarities to Go – there really are a lot of mini battles being fought all over the board. You can't let your attention wander and focus only one area or constantly chase weaker stacks around. Then there's the eternal question of whether it's better to make yourself stronger or your opponent weaker.

I'm sure there is a fairly optimum play style for those who want to analyse and take game seriously, but for more casual players I think psychology and tactical tricks are what will win the game. I was even thinking that despite the board setup being random, there are probably a few different openings that players can adopt. Tzaar has an abundance of tactical fun.

Anyway, back to game three. In this game I focussed on capturing her Tzaars, and pretty soon I had cornered her last Tzaar and won the game (much to my wife's disgust). This was the shortest game and I think that once we've a bit more experience under our belts it will be rare to lose a game by being reduced to 2 piece types from a quick attack. Fierce attacking may work on occasion as it puts you under a lot of stress, with having to check out so many different move possibilities. It might be fun to play with a timer.

So, all in all we both really enjoyed our first try of Tzaar and will play again as soon as possible. The nice thing about the game for me is that is gives me a Chess/Go fix but has very simple rules and a fast play time. I can't say it's a 'fun' game, as we barely spoke a word during all three matches (except to curse bad luck or taunt) but it really was intense and enjoyable. If you like abstracts and have a game buddy who doesn't mind some intense or nasty play, go ahead and pick up a copy of Tzaar. You won't regret it.

Image

BrenoK — 2008-07-11T03:36:19+00:00

(White's upper tzaar is one disc taller) It's black's turn, and his last tzaar is in serious trouble.

Thread: Re: review with some strategy insights

TheBrainDrainer — 2008-07-08T21:26:16+00:00

Indeed, some really good insights, thanks! Indeed, Tamsk is a nice game but doesn't really seem to fit into the series as well as the other games do. Unfortunately, I never quite mastered making multiple-capture moves in Zertz, and I agree that Dvonn has some counterintuitive elements in it. Until now, I liked Yinsh best, featuring a simple elegance and somehow, a higher fun-factor than Gipf. You should also try Punct, although it doesn't quite have the same elegance of simplicity as both Yinsh and Tzaar have. I was pleasantly surprised that so many elements of the previous games returned in Tzaar, which gave me the feeling that it could have been the last in the series following Punct rather than replacing Tamsk as the second game of the Gipf Project.

Image

trioker — 2008-07-03T22:28:03+00:00

T-Z-A-A-R: can you read it ?

Thread: Re: Encirclement

Alan Kwan — 2008-07-02T05:02:34+00:00

Has anybody else trying this strategy? My experience is that, it is very effective against novices; if you know how to do this, you will always win unless the novice starts the game by making two stacks right away.

Against an intermediate player, this is always a good strategy to play because it forces the opponent to make two stacks very early, removing his option of building a huge single stack: depending on the setup and opening, sometimes even building the Tzaar stack to 4-high could already be the losing move. And it doesn't cost you anything to try, unlike making double capture moves of Tzarras which will leave you in a weak position once the opponent defends.

Thread: Re: review with some strategy insights

Alan Kwan — 2008-06-30T03:25:19+00:00

My shop was closed because it was not making profit.

I'm now working as "rules director" for the World Series of Mahjong, and enjoying gaming with my family in my spare time.

Thread: Re: review with some strategy insights

ntrolls — 2008-06-29T18:30:06+00:00

Very insightful, thank you.

I still fondly remember the one day when I walked into the Tarot Games shop in HK a few years ago
Are you still running the retail shop? The webpage seems to be down. Hope everything is going well!

Thread: review with some strategy insights

Alan Kwan — 2008-06-29T17:24:28+00:00

I'm in love with TZAAR. Rated "11" for "exceptional" on a scale of 0 to 10.

Even if they don't publish it as a GIPF project game with Kris Burm's name on the box, even if they don't make the pieces in black and white with the good quality of GIPF series pieces, even if they paste some great or silly theme on it, you'll know that the game must have been designed by Kris Burm just by looking at the rules (available on the official website, so I'm not repeating them here), for you can see clearly its GIPF lineage:

GIPF's survival rule: in GIPF, if you lose all your GIPF pieces, or if you can't make a move (no pieces in stock), you lose the game. In TZAAR, if you lose all of all your pieces in one of the three types, or if you can't make a move (no possible move on board), you lose the game.

ZERTZ's and DVONN's shriking board: the board (playable spaces) shrinks as the game progresses, adding tension, strategic variation, and nicely controls the playing time.

ZERTZ's two-part move: in ZERTZ a normal move is "place a marble, then remove a ring"; in TZAAR a turn is "capture, then either capture again or stack up (or pass, but that's rarely used)".

DVONN's stacking: though in TZAAR it works very differently. The funny thing is, stacked pieces feel and function like GIPF pieces in GIPF (in fact they are a lot more powerful than unstacked pieces, than GIPF pieces are to normal pieces in the parent game).

DVONN's mobility: in DVONN, having more moves available in the endgame is an important strategic concern. In TZAAR, it's an explicit victory condition.

YINSH's movement: pieces and stacks move in a straight line (not restricted by stack height, hence closer to YINSH than DVONN). Upon second though, it isn't really the same as YINSH either, since you can't freely choose your move distance (have to stop at the first occupied space, not shorter).

And TZAAR also uses a form of capture similar to chess: you capture an enemy piece by moving onto it. But here your stack also has to be at least as tall as the captured stack (as in "open information" Stratego, or the Chinese animal version Dou Shou Qi).

TZAAR has borrowed some elements from previous GIPF games, and combines all the goodness into an exciting new game which is fitting to become the grand conclusion to the GIPF series. The previous GIPF games have been out for several years, yet Kris Burm is the only person who has the talent to merge them together to make TZAAR. While TAMSK is not "my cup of tea", I see it is a good game and a very interesting game design; yet I could only agree that, the decision to replace TAMSK with TZAAR is a good one, because TAMSK is an odd-man-out in the series, and TZAAR really belongs here. Unlike many of the decisions made in the game industry nowadays, this decision is no less an artistic one than a commercial one.

I loved DVONN. But that game has a big problem: the strategy is very counter-intuitive to the uneducated novice, in the same way as Othello. Although that game is won by piece count, the strategy is on mobility, and a tall stack has low mobility in general - that is counter-intuitive. In TZAAR, height means power, plain and simple. With a tall stack, I can chase my opponent's shorter stacks. Against a tall enemy stack, I keep my stacks well away (unless I can build enough height to chase it back), and try to cut it off from reaching my stacks and pieces.

Yet, TZAAR has its share of counter-intuitive strategy which a player must learn in order to stand any chance against an intermediate player who has learned them well. The victory condition is to eliminate one of the three piece types of your opponent, and this is the way in which the novice will lose most games. Ironically, when the novice tries to win this way by capturing an enemy Tzaar (the piece type which starts with the fewest number) at every opportunity, he will always be thwarted when the opponent stacks up his Tzaar before the last one is taken. The problem is that, on the game box is printed "on each move you may choose to make your opponent weaker (with a double capture) or to make yourself stronger (by stacking up)". This statement is certainly great for a sales pitch, and may actually have been in the original design conception, but in terms of strategy it is way off, for the novice who plays double capture moves to "make his opponent weaker" will lose every game. In order to win, you must make a stacking move at every opportunity, unless you have a clear reason (such as "I'll have a forced win after this") to make a double capture. This is the golden rule of thumb.

For the game strategy is not primarily focused on the three piece types; the game is focused on the stacks. You can win by eliminating an enemy piece type, but the last piece of your opponent's (if he is an intermediate player worth his salt) will always be on top of a stack, and in order to chase and capture it, you will always need your own stacks. Once you have a stack of each of the three types, your focus should be on your stacks, and the identities of your remaining single pieces will have much reduced significance (or rather, they are of but occasional significance). How close one is to defeat is not measured simply by the number of remaining pieces in each of the three types, but rather, by how vulnerable they are to being captured.

Then how does the novice player lose by a piece type being eliminated? It's because the intermediate player plays his stacks better. Against the obvious novice strategy of making one big Tzaar stack, the strategy of playing for "Tzarra encirclment victory" (http://www.boardgamegeek.com/thread/322463) is very effective, and to add to the puzzlement of the novice, (apparently contrary to the aforementioned golden rule of thumb) the game is finished by a series of double capture moves - from a forced win position where the novice's remaining Tzarras can no longer be stacked safely (and hence defenseless).

The next game, when the novice starts building a second stack for his Tzarra early, he finds that he will have to face the complexities of multi-stack play. Even given that one should be stacking by default every move, unlike the single-stack strategy there is now no simple and obvious decision: which stack should I build up, or should I start a new stack (and where)? The precarious balance this calls for is easily overwhelimg for the novice. One moment when he seperates his stacks, the intermediate player will send a big stack in between his two stacks, chasing them further apart and hunting down the more vulnerable one (a process which is often assisted by the novice's weak moves during his attempt to escape). And then the next game, when the novice combines his two stacks into a big one in order to defend against the same threat, the intermediate player immediately commences a deadly encirclement attack on the piece type which is not repesented by the combined stack. The key difference is that, the novice attacks blindly and thus will not succeed, while the intermediate player knows the opponent's weak spot and attacks it successfully (or at least gains an advantage by forcing the opponent to respond). A good player doesn't come into the game planning to follow a certain "strategy" (as some do in Princes of Florence or Puerto Rico), nor does he imitate the style of his opponent; rather, he responds to his opponent's play by exploiting any weakness he shows, focusing on what he stacks or does not stack.

The amazing thing is that, despite the novice player being repeatedly walked over, he is still enjoying the game and would not refuse to play again. This is the magic in this game: a game is fast, and each game the novice feels that he is learning something (thanks to the fact that the strategy is not too counter-intuitive), so he wants to play again because he thinks he can do better. The intermediate player, too, wants to play again, not for another walkover victory (honestly! ), but because he too is learning something.

I play and enjoy themed games; they are fun. And I once thought that abstracts are dated and boring. The GIPF project games have changed my mind with their elegant rules, their depth, and their quick playing time. In these aspects, few themed games (or even other abstracts) can parallel. And TZAAR is the masterpiece in the GIPF project.

Image

Carrotteer — 2008-06-27T20:06:00+00:00

Game in progress

Image

trioker — 2008-06-27T18:10:46+00:00

Homemade game

Image

trioker — 2008-06-27T18:07:37+00:00

Homemade game

Image

trioker — 2008-06-27T18:05:58+00:00

Homemade game

Thread: Re: Opening strategy

Alan Kwan — 2008-06-27T15:50:20+00:00

cymric wrote:
I disagree with your statement that the number of pieces doesn't matter.
I disagree with same statement, too. And I never said it. Read carefully.

What I said was, there is no reason to change the number of starting pieces between the players for the purpose of driving them to 'play differently', because the different positions of the pieces (relative to the players' stacks as the game unfolds) are already serving that purpose.

Now on to more strategy discussions.

It matters a great deal, or, more precisely, it becomes more important as more and more pieces are removed from the board.
While it does matter somewhat, a player's stamina (defense strength) in a certain type depends much more on the strength (height and position) of his stacks in that type, than on the number of single pieces he has on the board. Of course we should also take into account the possibility of putting a piece on top of one of your own stacks to change the type of the stack.

If I have a stack of Tzaar of four pieces in a safe position, losing my second last Tzaar is not a big deal. But if the three other Tzarras of mine are being encircled, losing the fourth remaining Tzarra can be fatal.

The loss of one stone out of nine or fifteen is not a big deal, the loss of one out of two or three is. You want the pieces to be as equal in numbers as possible because that allows you the greatest freedom of movement.
No. I want the stamina of my three types to be as equal as possible. (More precisely, I want maxi-min.) Stamina is a combination of "stack strength" and total number (and position), with the former carrying a much larger weight than the sheer number of single pieces.

The official website offers several hints, and I agree with them:

#1. watch your own stamina, but don't be too focused on your opponent's three types
- this is because defense (protect a type) is much easier than offense (eliminate a type). Beginners are not yet ready to plan a sophiscated, co-ordinated effort to mount a real threat, so getting too occupied with attacking will just ruin one's own position when the opponent easily defends against the attack by making a stack.

#2. make your Tzaar stack by stacking Tzaar pieces on each other
- Because by making a Tzaar stack (and planning to grow it further by a few pieces), I'm raising my Tzaar stamina greatly, so it doesn't need numbers as much as Tzarras or Totts.

After all, if you stand to lose the game because one type of stone is completely off the board, all your efforts must be directed towards avoiding that outcome.
And usually the most effective method to prevent defeat is to stack it up. It is much, much more reliable than having a few more single pieces (which will be picked off sooner or later, unless you stack them up). Even if your stack is only 2 pieces, there are only a few enemy stacks on the board which can threaten it. Compare to a single piece - the entire horde of enemy pieces can capture it any moment.

Of course, this does not preclude one's getting some single pieces or lowly 2-stacks into positional safety - that is the beauty of this game. But positional safety for single pieces is a late game thing.

Second, stacks are powerful, true. But you cannot afford to send them out gallivanting without some care, especially if they get within range of an enemy's stack---what you called a Zone of Control, I think. This checks the power of the stack.
This is not an American wargame in which combat is resolved by rolling dice. So when one moves a stack, one should know what he is doing.

It's kind of obvious, but then it's not really so. Basically, a taller stack can chase shorter ones, but when the players have multiple stacks, things can get complicated. For one, two smaller stacks can be combined to make a bigger stack.

In my games to date, the result was a few stacks held back well within one's own territory to avoid easy capture of rare Tzaars or the odd Tzarra, and one or at most two stacks on active duty. (Ideally such stacks would be Totts only---being the commonest and therefore easiest to sacrifce---but it's usually a Tzarra on top of a Tott, to gain some extra protection for the rarer piece.)

The biggest weak point of this game, now, is that there is no on-line play. Thus, players always play against the same opponents, and they fall into group-think. While it is of course important to protect your three types, the game has a lot more than that. Tzaar is really a positional struggle to get your big stacks into position to chase your opponent's small stacks, and small stacks to ensnare his single pieces. Generally one should aim to prevent loss of his stacks, and it is not too difficult to keep most of them alive; so once one has built a stack of each type, the focus should be on those stacks, while the number of single pieces of each type lying around is of greatly reduced significance.

Finally, if one player goes on a killing spree and allows the other to settle into a defensive stacking strategy, then in my experience the killer will lose. He simply runs out of things to kill, even though he will still have more stones in play! That sounds paradoxical until you realise that all of his stones will be singles, and those of the opponent stacks. And since the killer has more stones in play, the opponent can still rather easily make a hit on the first move of his turn. I lost a number of games this way until I realised the flaw in my approach: players cannot allow themselves to be too different in the number of different moves---hit or stack---they make.
Read my strategy article. One should usually make a stacking move each turn. One should ever make a double capture move only if one has a clear reason.

Though there are quite a few such reasons - the game is highly positional. But one thing for sure: if an enemy piece type has high stamina (a strong stack, or many pieces in stackable positions [not ensnared/encircled]), playing a double capture for the purpose of reducing its stamina is a wrong move, because you are not going to win easily by eliminating that type. On the contrary, you give up a precious chance to strengthen your position.

Let's get a bit extreme and say, white gets a handicap of either 5 (single) capture moves or 5 stacking moves(without any capturing) before black makes his first move. Which will you choose? I will undoutedly choose 5 stacking moves, and with that kind of advantage I'll win the game 100% of the time. 5 capture moves isn't as big and decisive an advantage: black will stack up his Tzaar on his first move, and you just can't take it out.

The result is that in my opinion, it doesn't make a lot of sense to formulate unwritten rules like 'if it is assumed that black will stack a Tzaar on his first move, then on white's first move, he should capture a Tzarra instead of a Tzaar'.
I think it's you who are making that kind of written rules first, isn't it?

First that in lieu of the importance of numbers, reducing the rarest piece to even rarer levels makes more sense (it's a victory condition, after all)
Merely reducing enemy Tzaars does not fulfill any victory condition. To fulfill the victory condition, you must eliminate the last one - which is expected to be well stacked up, hence usually difficult to eliminate. Especially for white. Going straight for black Tzaar elimination never works, since your stack begins half a piece lower and can never threaten the black Tzaar stack.

By attacking his Tzarras from the start, you force him to defend both his Tzaars and his Tzarras earlier. Why otherwise bother to force him to do something which he is going to do anyway?

and second, that it is still way too early to establish what kind of mode of play will unfold.
But you're making such predictions first!

In any case, we can predict that one will defend a type which has its stamina reduced. (Actually, that's the basic game theory assumption - not doing so is dumb.) And it is also reasonable to predict that, defense is quite easy when one focuses his efforts (especially for black!). Thus the most effective offense is to force the opponent to spread his defense thin.

In theory there is perhaps more strategy to this game than I first gave it credit for, but at the same time, I find it hard to translate these thoughts into practical and strong tactics which allow variation beyond 'keep close to your opponent in mode of play'.
The game is deep. And I have been working on, and will continue working on, putting the strategy into words. Please read my articles.

There are clearly stronger and weaker strategies. There is no reason to copy your opponent if his play is weak. If my opponent likes to pick on my Tzaars, I'll just let him, since I'm making a tall stack anyway. (And should I make a dumb mistake and lose that stack, I'm probably losing the game due to material anyway; having a spare Tzaar piece around isn't likely to make a difference in the end.) If my opponent makes quite a few double capture moves while doing so, I'll happily just keep building my stacks, and soon I'll have most of his pieces on the board ensnared. If my opponent doesn't see that I'm playing for an encirclement victory, I'll enjoy winning.

This game is similar to Go in certain ways. In Go, one doesn't play the last surrounding stone to actually take "dead" enemy pieces off the board; the dead pieces are dead enough, and it's a waste of moves to do so. Similarly, in Tzaar, it is much more useful to try to 'kill' enemy pieces by building stacks to ensnare/encircle them en masse, than to waste the stacking part of your move capturing a piece. Think about it - a capture move captures only one piece, but a stacking move ensnares a whole bunch of pieces. Of course, if I already have an encirclement victory position, that's another story.

Thread: Re: Opening strategy

cymric — 2008-06-27T08:03:22+00:00

I disagree with your statement that the number of pieces doesn't matter. It matters a great deal, or, more precisely, it becomes more important as more and more pieces are removed from the board. The loss of one stone out of nine or fifteen is not a big deal, the loss of one out of two or three is. You want the pieces to be as equal in numbers as possible because that allows you the greatest freedom of movement. After all, if you stand to lose the game because one type of stone is completely off the board, all your efforts must be directed towards avoiding that outcome.

Second, stacks are powerful, true. But you cannot afford to send them out gallivanting without some care, especially if they get within range of an enemy's stack---what you called a Zone of Control, I think. This checks the power of the stack. In my games to date, the result was a few stacks held back well within one's own territory to avoid easy capture of rare Tzaars or the odd Tzarra, and one or at most two stacks on active duty. (Ideally such stacks would be Totts only---being the commonest and therefore easiest to sacrifce---but it's usually a Tzarra on top of a Tott, to gain some extra protection for the rarer piece.)

Finally, if one player goes on a killing spree and allows the other to settle into a defensive stacking strategy, then in my experience the killer will lose. He simply runs out of things to kill, even though he will still have more stones in play! That sounds paradoxical until you realise that all of his stones will be singles, and those of the opponent stacks. And since the killer has more stones in play, the opponent can still rather easily make a hit on the first move of his turn. I lost a number of games this way until I realised the flaw in my approach: players cannot allow themselves to be too different in the number of different moves---hit or stack---they make.

The result is that in my opinion, it doesn't make a lot of sense to formulate unwritten rules like 'if it is assumed that black will stack a Tzaar on his first move, then on white's first move, he should capture a Tzarra instead of a Tzaar'. First that in lieu of the importance of numbers, reducing the rarest piece to even rarer levels makes more sense (it's a victory condition, after all), and second, that it is still way too early to establish what kind of mode of play will unfold. A lot more moves still await the players. It's like saying that in chess, 1. e4 should always be followed by 1. ..., e5. Well, that would depend---if I feel like surprising my opponent, I'll pick 1. ..., c5.

In theory there is perhaps more strategy to this game than I first gave it credit for, but at the same time, I find it hard to translate these thoughts into practical and strong tactics which allow variation beyond 'keep close to your opponent in mode of play'.

Thread: Encirclement

Alan Kwan — 2008-06-27T03:28:25+00:00

In DVONN, I talked about encirclement (surround some enemy pieces completely so that they cannot move). Now the concept is back in Tzaar, in a different form.

In Tzaar, a stack (even if only 2 pieces) has the power of ensnaring enemy pieces within its zone of control (one turn's movement range, i.e. double capture range). The ensnared pieces cannot stack safely because they will be immediately captured. In Tzaar, pieces which cannot stack are defenseless, and will eventually be picked off unless something is done to change the situation, such as escaping away from enemy ZOC, a taller stack coming to the rescue, or luring the ensnaring stack away with a gambit.

Ensnared pieces in a position with no easy escape and no prompt rescue are said to be encircled. In the early game, this is typically achieved by two or more stacks sandwiching enemy pieces with their ZOC: in that stage of the game, the board is quite full and long-distance movement options (for escape or rescue) are not easily available. Just like in Go (where pieces played near a corner can control the corner), sometimes a single stack can trap some pieces in a corner of the board, hence effectively encircling them, too.

Just like in Dvonn, it is possible for a strong player to play for an encirclement victory against a novice, although in Tzaar the encirclement is more subtle and even an experienced player may sometimes be caught in surprise. If the novice focuses on building a single Tzaar stack, the expert can systemetically capture those Tzarras near the Tzaar stack, while making several stacks near other Tzarras further away to ensnare or encircle them. If the novice doesn't respond promptly, soon (in 4 moves or so) he will find himself in a position in which all his remaining Tzarras are encircled by enemy stacks. At that point, unless he can find a way to save one of his Tzarras from the encircled position, he will lose the game quickly.

It is usually easy to defend against an encirclement victory if one notices the threat soon enough: just stack up a Tzarra in a safe place (while you still have a safe place). Especially, there are plenty of safe places near your Tzaar stack (before your opponent captures all the nearby Tzarras). Although good players won't lose to encirclement easily, a well-played threat can squeeze out some positional advantage.

At a later stage, encirclement can be used against Totts if one is not careful.

Thread: Re: Opening strategy

Alan Kwan — 2008-06-27T02:37:07+00:00

If it is assumed that black will stack a Tzaar on his first move, then on white's first move, he should capture a Tzarra instead of a Tzaar.

Maarten's criticism is missing something. Tzaar should not be a game in which you simply play one stack after another, i.e. make a new stack only when you're forced to. A stack, even if only 2-high, is incredibly powerful within a sea of single pieces, because opponent's single pieces cannot safely stack at all within its zone of control (one turn's movement range). So it is worthwhile to make more stacks, especially for white since his tallest stack will be behind in height anyway. A one-stack player, if he is careless, can lose the game to encirclement of his Tzarras or even Totts. There is no need to change the number of pieces of each player, because where they are at can be of critical importance.

It is difficult to catch a stack with just one big stack in the early game, if the opponent knows how to cleverly cut off your path of pursuit. (Though it is often advantageous to give chase since it puts pressure on your opponent.) Long-term strategy is very useful (i.e. you can play your early~mid game with the goal of stalemate win in mind); there is little point to chasing the immediately obvious goal if it relies on your opponent playing very dumbly to succeed. (This is the basic game theory assumption.)

Thread: Re: 3 games in a row with wife

Ambrose — 2008-06-27T01:22:22+00:00

Alan Kwan wrote:
While some players have compared Tzaar with chess, another subtle comparison is with Go. In Go, in the early game a single piece played in an empty area radiates some control within the area, and a few well-placed pieces can effectively secure control of an area. Similarly in Tzarr, a stack among a sea of single pieces radiates control in the area, and just two stacks of two or three pieces are often enough to encircle (effectively, kill) all the enemy pieces in the area.

Thread: 3 games in a row with wife

Alan Kwan — 2008-06-26T18:42:39+00:00

I grabbed Kaoru during a moment she's not too busy and convinced her to play Tzarr with me.

We use the random set-up. I place the pieces randomly and then let her choose sides.

In the first game, she took white. She proceeded to build a big white Tzaar stack, and then began singing "Amazing Grace". (For those who don't get this, there is a Japanese TV drama called The Giant White Tower which is about medical malpractice in a hospital. It uses Amazing Grace for its opening theme.) I stayed away from the giant white tower, and built several smaller stacks. I sneakily captured some of her Tzaaras (especially those near the giant white tower) and encircled (trap them with my stacks' ZOC so that they cannot safely stack) the rest. Before she knew it, Kaoru was down to three Tzarras, all of which were encircled. So it was a rather quick win for me.

Kaoru chose black in the second and third games (since I reminded her that black does have the starting advantage in this game). In the second game, Kaoru played better and survived until the endgame, but I had been gaining both material (captured a stack of 3 or 4 pieces, and a few 2-high stacks) and positional advantage. The endgame position was such that she had a big stack with no possible captures (only her own pieces in range; she got there capturing my Tzaara stack, but I still had two other Tzaara pieces left), while I had several smaller stacks in good positions (preying on her pieces and protecting my own). So I won when she couldn't make a capture (while I still had several moves left). (Actually, her capture of my Tzaara stack was a mis-calculation of mine; while I had a bigger stack guarding it, if I did capture her back, I would lose the game immediately since she could capture both of my Tzaaras on her next move. So I instead moved my pieces away and played for a stalemate win.)

In the third game, we soon built two stacks each. I combined my stacks to chase hers, and this forced her to combine hers too. She tried to chase me back, but she made the mistake of forgetting to stack once, so I chased her back. Eventually, she got herself cornered and lost.

I did make quite a few double capture moves in these games. In the first game, when Kaoru was over-building her one stack, I built several stacks to encircle her larger Tzarra groups, and then used a double capture move to pick off two remaining lone Tzaaras, and finally double captures to pick off the encircled Tzaaras (to quickly end the game before the giant white tower can come to the rescue). In the second game, a few double capture moves were made for positional reasons. In the third game, a few double capture moves were made in the chase action. So there could be in fact quite a few reasons one would want to make a double capture move in practice, and it is important that one does make the correct move when it presents itself. But the general theory remains correct: one should make a double capture move only if he has a clear reason.

While some players have compared Tzaar with chess, another subtle comparison is with Go. In Go, in the early game a single piece played in an empty area radiates some control within the area, and a few well-placed pieces can effectively secure control of an area. Similarly in Tzarr, a stack among a sea of single pieces radiates control in the area, and just two stacks of two or three pieces are often enough to encircle (effectively, kill) all the enemy pieces in the area.

Thread: Re: A perfect 10.

axelay80 — 2008-06-26T15:14:03+00:00

Freakin' awesome review. I definitely agree on the quality of the pieces, but as you said, they don't take away from the amazing gameplay.

Thread: Re: A perfect 10.

Ambrose — 2008-06-23T20:54:30+00:00

BrenoK wrote:
This really isn't much of a problem, though, since most of the time the newbie asks for a re-match. Tzaar is a fast game. Matches usually last 10 to 15 minutes. Half an hour is when it drags near the end (and those tend to be the best matches). One of the times I taught this it was even a little funny, the new player lost and just started setting up the pieces again, shuffling them around a little bit for a different setup, like he was just taking the action in his turn. He didn't even ask me if I wanted a re-match, didn't even look up from the board, it was like it was a natural thing that we played this game again. I, of course, loved this.

Go, tenacious newbies!!! You gotta love 'em.

Thread: Re: A perfect 10.

Ambrose — 2008-06-23T20:51:11+00:00

BrenoK wrote:
(I've sent a captioned picture of it to BGG, but it was denied because “it should've been sent under Creative”... WTF?)
I got that too. My sentiments exactly.

Thread: Re: A perfect 10.

Ambrose — 2008-06-23T20:49:05+00:00

BrenoK wrote:
The board in Yinsh is white, but it's not just white. There's a texture to it, a marble-like quality to it.
The texture is supposed to be clouds.

Thread: Re: A perfect 10.

latindog — 2008-06-23T19:43:05+00:00

Great review. I have only played a few games of Tzaar so far but have enjoyed all of them. In terms of the most chess-like of the Gipf games I would place it second to Zertz. Zertz is a tense tactical battle and one must look ahead multiple captures much like one does when forming combinations of forced captures in chess.

Thread: Re: A perfect 10.

BrenoK — 2008-06-23T18:06:57+00:00

By the way, Yinsh may have come off as a bad game after so many gameplay comparisons, and it was replaced by tzaar in the #5 spot in my top 10, but it's still a great game, a solid 9 out of 10.

Thread: A perfect 10.

BrenoK — 2008-06-23T15:01:55+00:00

I don't like the production values in Tzaar. Sure, it works perfectly (but how embarrassing would it be to make an abstract game with production issues?) and none of the issues actually hinder gameplay, but still... Yinsh is one of the prettiest games I own. It's an abstract, but its look gives the game quite an unique feel. Everyone I've played Yinsh with compliments how nice the game looks.

The board in Yinsh is white, but it's not just white. There's a texture to it, a marble-like quality to it. In Tzaar, the board is beige. Light beige on the inside of the playing space, dark beige on the outside. That's it. Like someone made it using paintbrush, happily indulging in the "fill it with color" function (the "spilling paint can" icon). Comparing it to DVONN is even worse, since GIPF's other "disc game" has a board with a very interesting "sandy" look to it.

The playing pieces leave a little bit to be desired. The pieces in Yinsh feel really good in your hands, very tactile. The material has an almost stone-like quality to them, almost as good as the ones in Hive. Tzaar just feels like plastic. Sure, the discs stack appropriately, no problems there, but I remember in a few times when playing yinsh a player in would just take a ring from the board and spin it in his fingers, like it would help him think his next move. And, you know, it kinda felt like it did.

I should mention that one of the white tzarras came with a gold smudge on its side (I've sent a captioned picture of it to BGG, but it was denied because “it should've been sent under Creative”... WTF?) I'll see if I can scrape it off with a knife or something, because it looks really off-putting. My copy also came with one extra black tott, which I keep under the insert so it doesn't mess with the setup of the game. I'm not a big fan of the “cracked paint” look the discs have.

Also, the box art isn't as interesting as the one in Yinsh. They've thrown in a "slick new frame" around the game illustration that, while doesn't actually look bad, makes the box look too different from the Yinsh one. This is a series after all. The game does come with a folder showing the box arts of the future GIPF re-releases (and, of course, they all have that frame, with slight variations on the color) but I liked better how the series looked before. They all had a color going for it. Dvonn was red. Zertz was blue. Yinsh, white, and so on... Now they're all enclosed in a frame, like a picture. The Gipf logo now is multi-colored, one color per circle, which sort of deviates from the aesthetics of the series . In every game you have black against white. It really is the classic abstract battle. Sure, Dvonn has 3 red discs, Zertz has gray orbs, but I think that's it. Why does the logo look all kiddy with these colors?

So, after all this, why is Tzaar a 10 out of 10? The game is fantastic. It's hard to elaborate much on that (unlike all the stupid nagging in the previous paragraphs), but I'll try anyway.

Unlike Yinsh, Tzaar does not wear its complexity on its sleeve. It takes a while to see all that is behind an apparently simple move. And unlike in Yinsh, I cannot give examples as to how the game actually develops to a newbie before the game really does develop. When I explain Yinsh, I usually say that a white ring in front of a row of four black markers is usually just as dangerous as if was in front of 4 white markers. It's easy. With Tzaar, things like that would be a lot harder to show before a match.

This really isn't much of a problem, though, since most of the time the newbie asks for a re-match. Tzaar is a fast game. Matches usually last 10 to 15 minutes. Half an hour is when it drags near the end (and those tend to be the best matches). One of the times I taught this it was even a little funny, the new player lost and just started setting up the pieces again, shuffling them around a little bit for a different setup, like he was just taking the action in his turn. He didn't even ask me if I wanted a re-match, didn't even look up from the board, it was like it was a natural thing that we played this game again. I, of course, loved this. The game has a compulsive quality to it, like Tichu, I can see myself playing many matches of this game over and over and over again. Yinsh was like that the first night I got it, but ever since I've reduced my plays of it to just one per game reunion, usually. Yinsh feels ponderous and slow in comparison, matches of it usually go for at least half an hour, depending on how seriously the players are taking the match. In Yinsh, the board changes a lot and quite visibly with each action, making each turn quite AP-prone, which doesn't really happen that often in Tzaar. I think for the next few weeks (or months) I will usually choose to play best out of 3 in Tzaar before opening up Yinsh.

Even though it's definitely not an obvious game, there are many intuitive elements to the rules. The concept of capture is known to almost anybody: who has never played chess, or checkers? Chess even has the positioning element that this game has, only with a lot less dynamism. The two moves per turn really helps keep things active. I've read here on BGG that this is the most chess-like game of the GIPF series, and from what I've played I definitely agree.

Only I think this game is much better than chess. Like I've already said, it's fast. Sure, add a chess clock and you have a much quicker match of the classic, but Tzaar doesn't need this. Every turn moves the game towards its end. Every move you make you take away one piece of the board, so the game comes with a built-in timer. There's even the impossibility of the draw: if you can't make a move, you lose, so sometimes you're fighting against time itself, while the opponent can afford to just dilly-dally, waiting for you to run out of moves. That is, of course, if you don't strike him where he didn't expect and you catch a surprise victory.

And that happens a lot. There are two winning conditions to the game, sure, but it sure does feel like it's a war on four different fronts. You can't run out of tzaars, you can't run out of tzarras, you can't run out of totts, and you have to be able to make your forced capture in your next turn. All this while trying to make your opponent fail in one of those conditions. It may seem like too much, and it even could be, but the result isn't an AP-filled game, but a very psychological duel. You can't attack in an obvious manner against a clever and experienced opponent. Sneaking up on somebody is a big part of this game, since there are so many options you can't possibly predict your opponents turn perfectly in most of the game, specially since there are two actions and they can be strategically combined to try to cripple you in your next turn: it's awesome to see the glint of a player's eye when they first figure out they can capture an enemy piece while aligning theirs in order to stack their tall tzaar in a good spot.

Aligning pieces is quite the strategic element in this game. Like I said, in every move you make, you move the game toward its end: you take away a piece from the board (either literally, in a capture, or by making it disappear, under a stack). You can't just move to an empty space, waiting for a next turn to unfold before you act. In fact, the manual even says this: once a space is emptied, it is never filled again.

A lot of games evolve into very fun fights. It's very common having a big tzaar stack chasing another, just wreaking havoc among the unstacked discs that were unluckily set up in their paths. A friend of mine even commented that it was like a fight between Godzilla and Mothra. Never thought you'd hear that from an abstract, huh?

Thread: Re: A Tale of Two Tzaar Virgins

board_games_reno — 2008-06-22T02:45:47+00:00

Strange...It made me wanna have a cigarette...

Fun stuff for great edition to the GIPF project.

Thread: Re: Kris Burm's gamebox description is inaccurate

Alan Kwan — 2008-06-16T14:30:06+00:00

Actually, the player who stacks up almost every turn is very likely to win against an opponent who tries to play some 'percentage' of moves as double captures. The opponent will have the burden of trying to capture some of the player's key stacks with his weaker height power on the board. If the opponent can't achieve that, the player will almost certainly win when the opponent can't make a legal capture eventually.

Thread: Re: suggested opening strategy

russ — 2008-06-16T05:20:52+00:00

Thanks for this and your earlier strategy article. These sorts of detailed game analyses and concrete proposals about game strategy are all too rare. (E.g. I had not previously thought of Black as having the advantage in Tzaar, but you make a good case for it. Now I wish I'd kept game records to look back on to see if either color seems to have an advantage in practice...)

You make some interesting points which I look forward to exploring when I next play Tzaar.

Thread: suggested opening strategy

Alan Kwan — 2008-06-16T03:56:15+00:00

In accordance with my other article, stacking up is by default the move one should make every turn in the early~mid game unless one has a clear reason for doing otherwise. In this article, I'll explain in more detail how and where the players should stack up in the opening.

Naturally, black should make a stack on his first move. Note that in this game, you don't necessarily have to start stacking with a Tzaar, because you can late convert your stack to a Tzaar stack by stacking a Tzaar piece on top of it. As long as your position is sound enough that you won't be blocked from doing so, you may even start with a Tzarra or Tott stack if it puts your stack at a better position.

On white's second move, of course he should also make a stack. Since his stack will be behind in strength against black's, he should build it in a safe location far away, so that he won't be easily chased by the taller stack. For example, if using the standard balanced setup, should black stack a Tzaar on a Tott in the center of the board, white should stay off the center and stack a Tzaar on a Tzarra (or a Tzaara on a Tott, etc.) on the opposite side of the board.

If white does make his stack far away, it is now up to black to decide whether he wants to make his first stack 3-high, or to start another 2-piece stack. The former option allows the black stack to outpower the white stack with its zone of control, and may even enable the black stack to chase the white one. The latter option enables black to cover a larger area with his zone of control, and creates the basic set-up for the typical two-stack (one Tzaar and one Tzarra) game. Here, black may want to make his stacks somewhat near each other, so that he can combine them should white decide to approach with his stack. Note that playing a double capture move at this point is not an option.

If white makes the mistake of putting his stack too close, black should by all means build up his stack and give chase. Even if you don't eventually catch the white stack [note 1], you'll typically gain some positional advantage in the process: with a higher stack, your zone of control nullifies that of a lower enemy stack in the same area. If your stack is only half a piece taller (i.e. same height at the beginning of your move), you should not fail to stack and let your opponent reverse the advantage.

The same does not apply to the player whose stack is half a piece shorter. If your stack is half a piece shorter, keeping building it single-mindedly will just give you a taller stack which is always still half a piece shorter than the enemy stack (unless you opponent makes a dumb mistake and neglect or fail to keep growing his stack against yours, which will not happen according to the basic game theory assumption). So don't do this, for it is the wrong strategy for the player with the lower stack. Instead, once you feel that your stack is at a comfortable height against the rest of the field (excluding the taller enemy stack), you should start another stack elsewhere. Note that playing a double capture move is still not an option: it is a last resort when your stack has to escape, or a positional move (not a material move) for cutting off the enemy stack's movement (i.e. you should have a clear reason for doing so).

By starting a second stack, white forces black into a dilemma instead of merely allowing black to grow his advantage indefinitely. If black continues to grow his tall stack, he is playing with one stack against two, and one stack has a smaller (although stronger) zone of control; while if black also starts a second stack, that stack will be playing against two stronger white stacks (and this is the reason why white might want to grow his first stack a few pieces first, although this is actuallly up to white and he may want to start splitting even at 2-high in some situations). Should white manage to cut off (prevent them from joining) the two black stacks, he may even be able to chase the black stacks by combining his stacks.

Thus, while black has a starting advantage in this game, it is not a large advantage (smaller than Punct, for example) if white plays smart (especially by staying away at a good distance to start with). With best plays from both sides, the game soon opens up into a battle with at least two stacks on each aide. (The game naturally encourages you to create at least two stacks, since you have to protect both your Tzaar and your Tzarra.) The battle is focused on the stacks, though; playing a double capture to aimlessly take out enemy single pieces is rarely a good move unless you can see a clear reason.

[note 1]: This game has a rule that helps the chased stack: the chasing stack cannot land on the space that the chased stack just vacated, because it is now a vacant space! So it is often difficult to catch a fleeing stack with just one stack: in the early game it has many ways to flee, while in the mid~late game it can cut off the chase path as it runs. But just like in chess, if you use two stacks to chase one, your chances of killing it are much better.

Thread: Re: stacking puts you closer to both victory conditions

Alan Kwan — 2008-06-16T02:12:39+00:00

Alan Kwan wrote:
Even a two-piece stack gives you a lot more options than a bunch of single pieces. This is because the two-piece stack is immune to capture by enemy single pieces. A captured piece presents no further options for you.

I feel like hitting myself for not having said this properly. The real power of a two-piece stack lies in the fact that, your opponent's single pieces cannot start stacking safely (without support or "guarding" by another stack of his) within your zone of control. Thus, having a stack in an area with otherwise only single pieces (a typical early~mid game situation) gives you the option to either further build the stack higher or build more stacks, while your opponent loses all such options with his pieces in the vicinity. Only a stack of at least two pieces have this power; any number of single pieces don't radiate this power with their zone of control.

Compare the following two situations:

1. I'm down to my last Tzaar, but I can safely start to stack it up this move.

2. I still have 4 Tzaars left, but they are all surrounded by enemy stacks' ZOC and cannot safely stack.

The second situation is much more dangerous than the first.

Thread: Re: stacking puts you closer to both victory conditions

Alan Kwan — 2008-06-15T18:25:23+00:00

An obvious counterargument is that when you keep stacking, you are reducing your own number of stacks, which brings disadvantages and reduced options (as well as advantages of fewer targets for your opponent). The balance of number of stacks vs height of stacks is debatable.

Even a two-piece stack gives you a lot more options than a bunch of single pieces. This is because the two-piece stack is immune to capture by enemy single pieces. A captured piece presents no further options for you.

I am not a one-stack player. I play a one-stack game only if my opponent takes the losing path of trying to overtake me in a 1-on-1 arms race which he is clearly destined to lose. (e.g. If I'm black, and white starts his stack too close to me, and I chase it, and while running white keeps building the stack in hopes of overtaking me, which is largely impossible since I can just keep stacking it up.) There are of course some advantages in building more stacks: more stacks have larger combined zone of control, and while you have the option of adding two stacks together when needed, you cannot do the reverse and break up a big stack. Also, creating a new stack reduces the number of weak pieces you have by two, instead of one when you merely make a stack bigger.

But by "stack" above I exclude single pieces. Single pieces are mostly good for the purpose of stacking. They have a strong drawback (among others) that, they assist the movement (and hence zone of control) of your opponent's (tall) stacks more than your own! This is because the two-part move rule allows only a capture move for the first part, so a stack can 'turn' and move on in any direction at an enemy piece, but can only either stack on a friendly piece, or the friendly piece can move out of the way to enable the stack to move further on a straight line only.

I fully agree with your point that near the endgame, the positional advantage is often more important than raw stack strength. At that point, even single pieces have the advantage that you can move them directly, and hence can shape the 'terrain' to your advantage more handily. For example, if a very tall enemy stack has no move anywhere except onto two pieces, if they are both your pieces, you can move them both away in one turn and isolate the enemy stack, while you cannot do the same if they are enemy pieces. (Above is another example of a clear reason for a double capture move - while you should make a double capture move only when you have a clear reason, if you do have a clear reason, you should not hesitate to make the move.)

Nevertheless, in order to reach the endgame stage, you must have used most of your moves in the early~mid game for stacking in the first place. If you keep making double captures for your first 5 moves, your opponent can make three or four stacks whose combined zone of control pretty much covers the whole board! If you don't stack up from the early get-go, you'll soon have no safe place to stack at, and will be quickly eliminated defenselessly.

If they also can't capture any other of my stacks (often due to stack heights, admittedly), then they lose.

When it is not due to stack heights, it is due to your having vastly reduced your weak pieces by stacking them. If you have lots of single pieces, a smart opponent (our standard game theory assumption) will aim his stacks at them, so you can't run him out of moves. Stacking not only builds strong stacks for yourself, it also reduces the weak pieces you have, hence reducing the zone of control and movement options for the strong enemy stacks. You can get an endgame positional advantge only if you follow the general advice of stacking by default unless you have a clear reason.

I mentioned above that even single pieces can have the advantage of flexibility in shaping the terrain in the late game, but that's for a few single pieces on the board. If you have huge seas of single pieces on the board (while being behind in stack material), the terrain is hugely unfavorable for you to begin with, and a little shaping isn't going to save you.

Thread: Re: stacking puts you closer to both victory conditions

russ — 2008-06-15T05:30:50+00:00

I agree with a lot of your analysis and that of course one needs to make some stacks. An obvious counterargument is that when you keep stacking, you are reducing your own number of stacks, which brings disadvantages and reduced options (as well as advantages of fewer targets for your opponent). The balance of number of stacks vs height of stacks is debatable.

But there is one significant other thing you've overlooked. You wrote:
What about the second victory condition? Imagine a game in which one player uses most of his moves to stack, while the other uses most of his moves for double captures. If neither player wins by checkmate, at the end we will reach the situation where the stacker has a few stacks left, while the capturer still has many single pieces. Unless the capturer can capture one of the stacker's stacks, the stacker wins because the capturer cannot make a move.
It is not quite true that to win via the second condition (opponent can't capture) you must have taller stacks, although that certainly helps.

You can win by the opponent not being able to make a capture, yet this is not due only to stack heights. E.g. my last Tzaar is only 2 high, but no matter what happens, they can't capture my last Tzaar no matter how we play it out, even if all their stacks are higher than my Tzaar stack! If they also can't capture any other of my stacks (often due to stack heights, admittedly), then they lose.

This kind of thing is not only theoretically possible but in practice I have seen it happen multiple times. For those who don't see how, consider it a puzzle. Solution below.
[o]Stacks can end up on points that have no other stacks on their lines; thus such stacks are effectively out of play and can neither capture nor be captured. Maneuvering a short stack into such a situation is a nifty efficient way to ensure that you no longer need to worry about stacking and protecting that type of piece. I've had several sessions where the endgame ultimately came down to positional maneuvering and reducing target stacks offered to the opponent more than stack sizes per se.[/o]

Thread: Re: Kris Burm's gamebox description is inaccurate

russ — 2008-06-15T04:11:48+00:00

Alan Kwan wrote:
I beleive that, in general, the player should usually make a stacking move instead of a double capture. See my strategy article.

Hence, the game box description is just a marketing phrase. Strategy-wise, it is inaccurate. The player who buys it at face value and makes lots of double capture moves will likely lose.
But a player who only makes 1 capture per turn will probably also lose. A mixed strategy seems best to me, and so I would not call the description inaccurate or mere marketing fluff. (It may be true that more turns should including stacking, but that just means that good strategy is not an exact 50/50 mix; but it would have been surprising if it was!)

Thread: Re: Kris Burm's gamebox description is inaccurate

Alan Kwan — 2008-06-15T03:24:27+00:00

I beleive that, in general, the player should usually make a stacking move instead of a double capture. See my strategy article.

Hence, the game box description is just a marketing phrase. Strategy-wise, it is inaccurate. The player who buys it at face value and makes lots of double capture moves will likely lose.

Thread: stacking puts you closer to both victory conditions

Alan Kwan — 2008-06-15T03:06:29+00:00

White starts the game by capturing a Tzaar. Black makes his revenge by capturing two white Tzaars.

Black has just made a wrong move.

According to the rules, for the second move of your turn, you have three options: capture again, stack, or pass. But actually in most cases, one of these three options would put you closer to victory than the others.

Let's begin by studying the victory conditions:

1. You win by capturing the last emeny piece of any of the three types.

2. You win when your opponent cannot make a capture at the beginning of his turn.

Here we use the standard game theory assumption that your opponent is playing a competent game, i.e. he doesn't throw the game with a very stupid move. Thus, you should plan your moves against your opponent's best moves. If a player makes his moving hoping for novice's mistakes from his opponent, novices are all he will win against. No good Shogi player bases his play hoping for his opponent moving into check (even though the rules of Shogi don't prohibit winning that way).

Under this assumption, you can expect that your opponent will stack up his last piece of a threatened type. Thus, in order to capture that one, you must "checkmate" it with your stack(s) which are of equal or better strength. If your opponent starts stacking before you, you will be tracing in stack strength, and you will be the chased, not the chaser. If you hope to win by the sudden death condition, your biggest hurdle is to have enough stack strength to outpower that last stack. Taking out the other, earlier, lowly single pieces of that type is relatively trivial. OTOH, if you hope to avoid losing, you have to stack up to defend your last Tzaar or Tzarra from the vast sea of single enemy pieces.

What about the second victory condition? Imagine a game in which one player uses most of his moves to stack, while the other uses most of his moves for double captures. If neither player wins by checkmate, at the end we will reach the situation where the stacker has a few stacks left, while the capturer still has many single pieces. Unless the capturer can capture one of the stacker's stacks, the stacker wins because the capturer cannot make a move. A stacking move creates or consolidates a strong stack for you, and eliminates one or two of your single pieces - which are easy captures by your opponent.

At the beginning of the game, the board is full of single pieces. Any of these pieces are easily captured by any other piece which can move onto it. OTOH, once you stack two pieces, that stack becomes immune to capture by any single piece, which are most of the pieces on the board (in the early~mid game). Stacks are where your real power is; single pieces are mostly just terrain to fight on until you start stacking them.

Of course, there are situations where you clearly should make a second capture move. Some of them are:

1. You win instantly with that move, or have a forced win (checkmate) from that move
2. You are capturing a stack, or can checkmate a stack
3. One of your stacks is escaping from capture
4. You can force your opponent to use his next move for a double capture in order to escape

... and so on. If you have a clear reason, you should of course capture again. But without a clear reason for otherwise, you should be using most of your turns (in the early~mid game) for stacking, because that brings you closer to both victory conditions. Making a double capture brings you closer to the sudden death win, but in a smaller degree (in general) than stacking does. And it brings you closer to defeat by the second victory condition. Thus, you should not give up every opportunity to stack up unless you have a good reason.

This is not to say that the option of making a second capture is not important. It is a very important rule, because it defines the "zone of control" for a stack: with a double-capture move, a stack can "make one turn" at a spot occupied by an enemy piece, or a friendly piece can be moved out of the way to un-block the path of your stack. Your opponent cannot safely start stacking within or move a weaker stack into your zone of control (unless he has a stronger stack guarding the spot, just like in chess).

I'm not saying that single pieces cannot be valuable: they are your stacking material, and a spare surviving Tzaar or Tzarra gives you some flexibility with your main Tzaar or Tzarra stack. What I'm saying is that, it is usually not worth it to give up your chance to stack up just to take out another single enemy piece.

Thread: Re: Kris Burm's gamebox description is inaccurate

russ — 2008-06-10T18:27:56+00:00

htjester wrote:
He's such a great game designer that I've got to believe you'd be quite pleased with Tzaar.
True true, he's got a talent for game design. This past weekend I played Quads (which I'd never heard of before) because it happened to be on the table at a game con, and I was pleasantly surprised by it. I looked it up here at BGG later and discovered it's a Kris Burm design from 1996!

Thread: Re: Kris Burm's gamebox description is inaccurate

htjester — 2008-06-10T18:12:31+00:00

Balthus_Dire wrote:
I haven't gotten TZAAR, and I'm not interested in doing so-- it seems so close to DVONN.

Well... not exactly! Tzaar has stacking but it's really a position and capture game. I'm very intrigued to play more. (Now if I can only find Gipf nuts in the Chicago area.)

I haven't played Tamsk, and have declined to buy it because the buzz is that it's not a brain-burner (my preference), it costs more, and most importantly, Kris Burm himself has been trying to develop more consistent sand-timers. It does sound wonderfully novel and fun.

Kris Burm's consistent advice has been play his games in any combination to your liking, including with other games. He's such a great game designer that I've got to believe you'd be quite pleased with Tzaar. Remember, you can always try it on a Gipf board with 15 basic pieces as Totts, 9 Zertz potentials as Tzarras, and 6 other potentials as Tzaars.

Tim K.

Thread: Re: Kris Burm's gamebox description is inaccurate

Balthus_Dire — 2008-06-09T18:21:55+00:00

What I love so much about the GIPF project is the lack of uniformity, yet the strong connection feeling-wise. I haven't gotten TZAAR, and I'm not interested in doing so-- it seems so close to DVONN, and to be honest, I haven't forgiven it for bumping TAMSK, which I feel was an egregious error. In my mind the series was complete before TZAAR, and I won't be participating in any switch-out of PUNCT either, should that happen. I may later buy TZAAR, etc. as separate games, but not as part of the long ago completed GIPF project, which I fear Burm may be convinced to perpetually fuss with, to the project's detriment.

What I love most about the GIPF project is that in the 'potential' game I have an arsenal, a toolkit of very different brain-style games to pull out against different opponents--done in the way that I feel will be most advantageous against my opponent. Do I need to push a TAMSK potential now, or a DVONN? So in this way it delights me that the lightning speed of TAMSK is so very different from the slow sweaty wresting match that PUNCT is, and worlds apart from both is the trance music-like futuristic smoothness of DVONN.

We play with Burm's variant that once the # of potentials is decided (6, 9, 12, 15, or 18) each player can choose separately the amount of each potential that she will be playing with to make up the decided total potential amount. I have two sets of each potential to facilitate high variation. This makes for a fantastically fun game. My buddy Jonathan is fantastic at TAMSK, has the mind for speed, but I can wrestle him to the ground in PUNCT, so commonly we'll each load up on the other's weakness, but you never know when someone will throw a curveball and take 6 ZERTZ because they've done some study there.

My ranking, favorite to least, is DVONN, TAMSK, PUNCT, YINSH, ZERTZ, GIPF. Although GIPF is last, I understand that it is the only one which can hold all of the other games, so that ranking only makes sense if each game is viewed separately. The project as a whole is, of course, stellar.

Thread: Re: Kris Burm's gamebox description is inaccurate

Ambrose — 2008-06-09T17:34:29+00:00

htjester wrote:
BTW, I wasn't previously aware of the tournament version of Gipf... I will definitely be trying it out.

Whah? No wonder!!! The tournament version is the only version you should play!!! No wonder you didn't like it so much. With the tournament version, you get a feeling of "resource management"... in an abstract! No fooling!

Thread: Re: Kris Burm's gamebox description is inaccurate

htjester — 2008-06-09T17:30:04+00:00

Thanks to everyone for your feedback. BTW, I wasn't previously aware of the tournament version of Gipf... I will definitely be trying it out.

Image

BrenoK — 2008-06-09T14:13:01+00:00

Cracked paint on the discs in my copy... Why didn't they just use stickers?

Thread: Re: Opening strategy

cymric — 2008-06-08T18:47:48+00:00

Personally, I don't feel that there is that much strategy to this game. It plays quick, and is fun enough for a rematch or two (or three, or four, ...) but I've not seen much variety in the methods used to win from an opponent. Invariably, the game begins with a few tzaars which are killed off, and then the remainder is put into stacks so they are not that easily taken. (This is undoubtedly the reason for the fact that White begins with just a single move.) Then it's either chasing big tzaar stacks around which are growing steadily, or knocking the tzarras off the board, or a combination of both. Then in defense the tzarras are stacked, and once you reach that phase the game is beginning to devolve in a battle to still have sufficient options for making a kill once it's your turn. There's plenty of tactical variety in setting up little traps for your opponent and causing him/her to lose sente, but strategically, I doubt there's more to TZAAR than the above. Fortunately, the tactics are still fun to work out, and the game is quick enough that to date I've not found it a chore. But I do believe it will hurt TZAAR's replayability in the long run.

I think the game would gain more strategic variety if White and Black had a different number of tzaars, tzarras and totts: the players would be forced into different strategies straight from the beginning while still pursuing the same goal. Now one player almost has no choice in doing what the other player does, simply because the balance is lost otherwise. Unfortunately, changing that amount raises questions about the minimum number of stones which must be removed before the game is won---one player would always be on the defense, and the other on the offense. In the end, that's not much of a change either .

Thread: Re: So, are we going to see this on Boardspace.net?

russ — 2008-06-07T03:49:35+00:00

ddyer wrote:
My copy will arrive tomorrow. The omens are good, but it's not definitely scheduled yet.
Good news indeed! Tzaar is my favorite in the Gipf series. Actually it's become one of my favorites overall. So I certainly encourage you to implement it at boardspace.net...

Thread: Re: Kris Burm's gamebox description is inaccurate

BrenoK — 2008-06-07T02:28:44+00:00

Heh, nice Faidutti quote.

Thread: Re: So, are we going to see this on Boardspace.net?

ekted — 2008-06-07T01:56:50+00:00

The bots will be tough to beat for this one.

Image

Suzan — 2008-04-25T11:29:14+00:00

box back new edition

Image

Suzan — 2008-04-25T11:23:56+00:00

Box front new edition