Instead of making all of the dice available to every player all the time, I wanted to introduce a scout action where the player would randomly select X number of dice from a bag that they have scouted. This creates a "do the best with what you get" scenario. This created die type #1: Scouts. Players keep scouted dice, but cannot roll them until they add them to their society.
Next, I wanted to give the player two ways of adding scouted dice to their society. At first, I was thinking of using the war/peace diverging paths, but it didn't make sense to me that you could forcefully add a person to your society. Instead of ditching the military option, I decided to make the two paths to adding more dice "provide protection" and "recruit". Right now I'm giving recruit a religious bent, so two more types: Militia and Priests.
Finally, I wanted some modifiers on your typical actions, but I didn't feel like one more type would provide enough strategy space. Again, two paths to modifiers seemed interesting to me. First I wanted to be able to enhance a particular type of your choosing, so I decided to introduce small tech trees for each type. This tech tree could be advanced with the use of research; type #4: Engineers.
Finally, I wanted to allow modifiers that weren't necessarily tied to a certain type, so I introduced buildings; type #5: Builders. While Engineers develop a tech tree (where some technologies have prerequisites), Builders can build anything they can afford.
The general purpose of both Builders and Engineers is to introduce game elements that create exceptions to the rules.
With the 5 roles comes 5 actions: Scout, Recruit, Protect, Research, and Build. The roll of the player's dice determines how much support they have for performing a particular action that turn. Since the purpose of using dice is to introduce a random element, not every face on a given type's die should display the type's corresponding action (e.g., a Scout die should not have 6 Scout icons).
First, each die will have 1 - 3 or 4 (or 5?) VP icons. The total number VP icons on the die represent the die's value at the end of the game. However, the VP icons do not support any action. Therefore a die that is worth more VP at the end of the game will be less likely to help you build your engine. A die that has more icons has a better chance of helping you to build your engine, but it will be worth less points at the end of the game.
Next, a given type will actually contain icons for two actions. The majority of the icons will always be the primary action that corresponds to the given type. However, they will also contain icons for a secondary action. For example, a Scout's primary action is Scout, and its secondary action is Protect. A 1 VP Scout die will contain 3 Scout icons and 2 Protect icons (and 1 VP icon). A 2 VP Scout is 3/1; 3 VP is 2/1; etc. This allows for more variability that the player has to manage, while also providing access to actions without procuring the type of die that has that action as its primary action.
A turn would look something like this:
- Active player rolls his dice
- Active player chooses his action
- Other players roll their dice
- Active player carries out his action
- Other players choose
- carry out same action
- task a die
At this point I believe that the active player should choose their action for the turn before the other players roll their dice. This is mostly to keep the game moving and mitigate analysis paralysis. However, if it makes for better choices, I am open to changing it so that everyone rolls before the active player chooses the action for that round.
In each turn, if the an inactive player does not want (or is unable) to perform the action that the active player chooses, he can instead "task" a die. This simply means that he can take any of the die that he has rolled that turn and set it aside, locking in its current value. On a future turn, he can choose whether to keep the current value of any tasked dice, or re-roll as many of them as he wishes. Once a player decides to perform an action, all of his tasked dice must be rolled on the next turn, regardless of whether they are used or not.
That's the game in a nutshell. Players will spend each turn building their engine by adding more dice, adding buildings, or researching new technologies. Currently I think the game-end condition should be when a player has no scouted dice, and there are no more unscouted dice (the bag is empty). Players tally their points from their dice and buildings. The player with the most points win!
Once I mock up a player board (maybe multiple -- with each one giving each player a different set of dice to start and/or a special bonus) and determine the dice distributions (roughly symmetric across die types, with low-value dice being more common than high-value), I'll have to start on a prototype.
I guess I need to find a bunch of dice and stickers...