The windmills of my cards

The deal description is wrong.

The first player should get the 1/5/9 in one suit, the 2/6/7 in another, and the 3/4/8 in the third. The second player should get the same thing but with the suits allocation rotated once. The third player should get what’s left, which is the same distribution with the suits rotated the last time.

I’ll get a formal versioned PDF of the rules put together and posted soon.

The reduction of bouncing water

‘Ohana Proa did not make the cut for Hippodice. It scored 3.33 versus the required 3.1.

I’m pretty satisfied with the the game as-is. The rules are in good shape, it plays well, I enjoy it. Perhaps it is time to start talking to the Peter Eggerts of the world.

Onward ho!

Boil and bubble

Initially targetted as a perfect and certain information three player card game with both a high Take-That! factor and high control. This game is the product of roughly 15 minutes actual thought and a little mumbling to myself while taking a shower.

The game consists of two decks of cards. The player deck and the prize deck. The player deck consists of 31 cards: three suits of 11 cards (red, green, blue) with values ranging from 1-9 inclusive, plus three I Win! cards and three blank player identification cards. The prize deck consists of nine cards, three in each of the three player suits with values of 4, 6 and 9. The back of each prize card is coloured to match its player suit.


  1. Shuffle the prize cards and deal them out in a face down line
  2. Pick a start player and give them the red player identification card and the red I Win! card. The green player identification and I Win! cards go to the player to their left. The blue player identification card and I Win! cards go to the player to the start player’s right. Each player publicly displays their player identification card and takes their I Win! card into their hand.
  3. Separate the three player suits and sort the cards in each suit into ascending order.
  4. Starting with the start player deal one red player card to each player until the deck runs out.
  5. Starting with the player to the left of the start player deal out the blue player deal out the green deck in similar fashion.
  6. Starting with the player to the right of the start deal out the blue player deck in similar fashion.
  7. The dealt player suit cards form each player’s hand along with their I Win! card.


  1. Turn over the leftmost prize card and add it to the prize pile (beside the row of face down prize cards)
  2. In turn order starting with the start player each player plays a card face up before them.
  3. Resolve the trick.
  4. Repeat from step #1 eight more times, starting with the winner of the prize suit in the last trick
  5. Each player claims half their prizes
  6. The player with the highest score wins.

Resolving the trick:

The trick is resolved by resolving each suit in order, starting with the suit of the prize card, then the remaining suit with the highest value card played (tiebreak goes to the player suit to the prize suit player’s left), then the last remaining suit. If no cards are played in a given suit, then that suit is not resolved for that trick. If no card is played in the prize suit then the other suits are resolved as necessary and the prize card is left in the prize pile. In this case the lead moves to the next player to the left of the previous lead.

Resolving the prize suit:

  1. Whichever player played the highest valued card in the suit of the prize card takes the prize pile card(s) and places them along with the card they played in their score pile. If any player played their I Win! card then they automatically are considered to have played the highest card in the prize suit. If multiple players play their I Win! cards then the earlier played cards are considered “bigger” than the later played cards.
  2. If another player also played a card in the same suit, then the player from #1 separates their score pile cards into two face up stacks. The player with the next highest card in the prize suit then takes their choice of the two stacks and places those cards in their scoring pile along with the card they played.
  3. If the third player also played a card in the prize suit, then the player from #2 separates their score pile cards into two face up stacks. The player with the smallest card in the prize suit then takes their choice of the two stacks and places those cards in their scoring pile along with the card they played.

Resolving other suits:

  1. The player identified by the suit divides their scoring pile cards into two piles. The player of the highest card in that suit takes their choice of the two stacks and places those cards in their scoring pile along with the card they played. If the player played a card in their own suit it is merely added to there score pile.
  2. Likewise again for the next lower card in that suit.
  3. And possibly yet again for the third lowest card in that suit.


Each player sums the value of the cards in their scoring pile. I Win! cards are worth nothing. The largest score wins.

Variant #1:

When dividing their scoring pile into two sets for another player to choose from, the splitting player must ensure that the two piles have as close as possible to the same number of cards as each other.

Variant #2:

A tenth trick is played, but without any prize cards. The suits are resolved starting with the suit of the player that lead the trick.

Decompositional question

A night of odd dreams.

Let’s say I adopted a hyena model. The hyenas would simply be autonomous critters that wandered the landscape consuming corpses.The pack would grow as it ate, and when larger than X would split into two packs. A pack would consist of N hyenas per tile (not more than two), and would move deterministically toward the nearest corpse, injured or healthy player token (in that order) that another pack is not heading for, twice a day. If more than Q hyenas stepped on a mine it would explode (no choice for the MP) with mostly standard effects (less damage for the hyenas or hyenas heal over in 2 turns).

The real purpose of the hyenas would be to provide an additional drain on the MP’s deployed munitions. I fear that without such a drain the map would progressively fill with munitions until every tile was loaded by the end of the game. Additionally the the hyenas would provide a distracting target for the MP: the MP can’t afford to let the hyenas run about uncontrolled. Finally, the hyenas could provide an interesting set of tactical opportunities ala: 1) Get kids eaten by hyenas by front and back edge of map (not easy, but bear with me), 2) follow hyena pack back across landscape in safety, letting the hyenas detonate any mines along the way.

A possibility. However if I adopted the hyena model, how would it also translate to the disease attack themeing? If the players are diseases attacking a host body and the MP is the host’s immune system, what would the hyenas represent? Surgical excision of dead material? Hurm, there are all sorts of interesting things that lead from there…

Numbers are next, then rules, then simulation.

Fearing fungal dismemberment

I was dreading making the tiles for this game. ~150 Carcassone-esque tiles with adjoining features that form a coordinated land image? The prospect of drawing, printing, mounting and cutting such tiles with any hope of reasonable registration or attractiveness of final result was not enheartening. Urk, no thanks, not me.

A recent trip to the local craft store (D&J Hobbies in Saratoga) gave another idea: Wood squares (roughly 5cm square) painted basic green using poster paint. Different shades of green could indicate plains versus forest. Gray or gray spatters for rocks. Blue or blue strip for river/pond/lake. 150+ tiles should take less than an hour to produce. The map can then be made by simply stacking the tiles into a pleasing form, the resulting grid forming the play-map, the elevations and contours made providing some of the terrain interest for the variant move rules.

Possibly even more pleasingly the smell of the poster paint is reminiscent of preschools. Something is begging me to do this game.

The virus model for the game theme is also growing on me. Becca (a local) has been complaining vociferously that viruses don’t have intelligent direction and that their mutation rates are too low to support the other aspects of the game design, and … she’s right. I suspect mostly however that she’s really just likes the Evolutionary Psychology aspects of the infanticidal theming. I have sympathy. Still, I do like the idea of players playing the role of plagues fighting to infect a host body (represented by the Military Player). It is H G Well’s War of the Worlds all over again.

Meanwhile the repeated requests for roving packs of corpse eating hyenas aren’t exactly falling on deaf ears. Certainly I can imagine interesting models which tie their breeding rate to the food supply via player corpses and then allow fully automatic movement and manipulation of the hyenas (cf the spiders in Atta Ants), but while interesting that also feels like it heads away from the heart of the resource-management design. Nice ideas, shelved for now.

Battle of the cleavage - splitting hairs

A jumble.


  1. a shared board across which the families move and a private board the Military Player (MP) maintains to track his own mine placements.
  2. every player has a private board and merely calls out the locations of their pieces as they move

Intuitively I prefer #1 as it allows for easier coordinated /shared action among players, as well as reducing the temptation and ease of cheating.

In mucking about in the area it struck me that a Battleships board would be perfect for the MP. They could use the same board and pegs, perhaps with coloured tips, to mark their mine locations secretly from the other players. It would only be a tracking aid however.


Players build a 10x10 landscape using square tiles. Landscape features include a river, forests, craters etc. Some features span multiple tiles. The total number of tiles in the game should probably be in the 120-150 range to allow for more variant setups.

Some terrain restricts movement. Some restricts movement for injured only (hard to crawl uphill when you have no legs). Possibly also movable objects (eg big logs) Rivers automagically move contents one tile in direction of flow.

MP seeds the land with face down activity markers. Some will be mines of various types, some will be nothing at all. In seeding the map it will be clear to the players how many munitions of each type were placed, just not where they were placed. The MP would record their munitions placement on their Battleship board.

Players start moving kids across the landscape. Movement is orthogonal. When a kid enters a tile the MP may declare that it detonates. If so appropriate results occur. Some detonations will may also detonate adjacent tiles. Appropriate results again.

Once a kid has entered a tile the tile is “safe” until the tile is left empty. When entered again it may detonate. Thus a kid may secure a tile and hold it secure while other kids hold adjacent tiles secure while a chain of kids traverses them.

Max population of a tile is limited. ~3 or ~5 probably.

MP weapons:

  • Standard mine. Kills tile contents.
  • Big mine. Kills contents, causes orthogonal neighbours to detonate (no choice).
  • Small mine. injures tile contents.
  • Cluster mine. Kills tile contents and spreads 1/2 kill mines to N of 8 surrounding tiles.
  • Mortar. Placed on back of map. Can be moved during nightly restock. On any patriarch action can fire. MP specifies target my placing up to two 3x3 grids of target markets on the board within N squares of mortar location (N=4?). After patriarch’s next action mortar lands. MP reveals targetting tile to indicate target. Mortar kills tile contents and injures orthogonal neighbours. Mortar causes target tile mine to also detonate with possible chain reaction for big mines.
  • Barbed/razor wire. Placed on a tile during nightly restocks. Reduces movement by healthy to 1/2. Injured can’t penetrate. Destroyed by mines and mortars.

A game is organised into days.

  • Each day starts with the MP buying munitions and seeding the field some portion of his stock. May also move mortar emplacements.
  • Then the patriarchs take turns in rotation, each choosing one of three possible actions. Basic actions are: Move, Queue and Equip (?).
  • A given action may only be selected so many times before it is unavailable. Then only the other actions are available.

Once two actions are depleted the day ends.

  • MP is paid cash for kills/injuries and MP may buy new munitions with cash.
  • Patriarchs:
    • Feed their at-home population
    • Unfed population starves and is injured
    • Excess food suppresses breeding rate and cash flow?
    • Receive charity food based on injury/kill rate etc.
    • Breed as a function of food supply change
    • Receive money from already escaped kids

New day starts.

Basic problems:

Patriarchs must:

  • ensure sufficient injury rate for themselves to sustain food supply for their population.
  • ensure sufficient population (inverse function of first derivative of food supply), to maintain activity
  • escape enough kids across field to secure money flow to equip later kids well enough to escape military build up
  • keep MP’s score within winning bounds
  • ensure MP income is low enough to ensure future surviveability.

MP must:

  • distribute income against munition types and expenses
  • distribute munitions effectively
  • detonate munitions efficiently versus score and income
  • keep his own score and specific patriarch scores within winning bounds.
  • generate future income for munitions

It doesn’t feel like it is holding together yet. There are kernels there, but nothing coherent yet. It is still incoherent. Part of the problem for me is that I’m trying to do trinary relationships for the first time. In comparison with ‘Ohana Proaexternal link which strictly implemented binary current translations pairs (A->B), I’m attempting trinary relationships with Splatter my Children and am finding it hard. The intent is for every decision to involve tradeoffs on not just two but three fronts.

I think it is time to get this thing out of my head and either make a quick and dirty slips-of-torn-paper prototype, or actually go full hog and write a little software implementation in Python.

Stretching nerves, render unto efficiency

One of the things about the limited action system (stolen from Wabash Cannonball, tho it is not original to there), is that it provides an interesting turn tension. The is functionally organised into rounds. Patriarchs rotationally take turns until they have exhausted two of the three possible action sets. Any action used by a player which isn’t moving a kid across the minefield is essentially a wasted action that gets less done before the military player gets his next restocking of munitions. But without using those other actions players can’t get kids onto the field, can’t breed new kids etc etc etc, all the necessities required before being able to move kids across the minefield.

This is cute: the more you waste time with getting ready to move kids across the field (which may enclude equipping them with anti-mine devices like chains and brooms), the more impenetrable the minefield will be. The less time you spend on those background actions, the more readily and easily your (expensive) kids will be offed.

The result, I think, will be encouraging tight coordination between relatives. A player may move their kids or their relative’s kids (half distance?) on their turn. Thus the players may attempt to optimally distribute kid-moving versus non-kid-moving actions among themselves so that the maximal distance is covered and the least possible number of non-kid-moving actions are used as required for each player’s success. Of course the fact that the relative linkage is leftward will make that interesting: you can move your kids or the kids to your left, but the leftward player cannot move your kids, only his and his leftward player’s kids. The result is a leaning domino chain of delegated responsibilities.

Hurm. It may be more interesting to have the relative relationships bind in the direction opposite to turn order. Must think about this. The short version is to pick the direction which is more difficult to manage.

Odd thought: Returning an injured kid back to the home side of the board generates extra bonus food? (Display before media? Extra action choice?)

Wages of sincerity

Knowing me, a part of the system will turn into a currency management problem, effectively a question of how to process children into VPs, so I might as well confront that aspect now. The obvious elements:

  • Healthy and injured family (injured generate more food), healthy and injured relatives (injured generate more food for them and less for you) and healthy and injured non-relatives (injured reduce food allocation for you and your relatives)
  • Population growth rate of your family and your relatives (function of what?)
  • Death and injury rate of your family and relatives
  • Kill and injury rate for military player (higher rates produce more replacement munitions faster)
  • Family population versus food allocation (starvation), and ditto relatives
  • Population camp versus population traversing minefield versus escapee/saved population
  • Military player’s score and patriarch’s isolated and combined score.

Other thoughts:

  • Possible distribution of sexes? Female score more, or injured female generate more food, or other sex-tied relation?
  • Cannibalism
  • Lotteries
  • Excess food discarded or increased mortality/injury rate due to over-eating? If injury then there’s a nice feedback loop where excess food produces injuries which produce more food.
  • Trade system for giving food to other players? How would the value of the transaction be recorded? VPs? Suppression of next turn’s food generation?
  • Increase breeding rate whenever food production falls. Decrease breeding rate whenever food production rate rises. No change on stasis?