Saturday, December 05, 2009

Dominion engine

Dominion is arguably the most innovative game to hit the hobby game industry in a long time. It appeals to a lot of people for a variety of reasons. For me it invokes the feelings I used to get when building decks for Magic: the Gathering. In Dominion, turn by turn you modify the contents of your deck, adding better action cards and getting rid of the less efficient starting cards. Lately I've been pondering the use of this "engine" to promote a game...

Last night I played the recently released Thunderstone, which is an RPG style adventure game which builds on the Dominion engine. In Thunderstone you "play Dominion" in order to fill your deck with heroes, weapons, spells, and equipment and then go dungeon diving to defeat monsters. It's an interesting extension of the Dominion mechanism, but to me it felt like a lot of work for what is largely just playing Dominion.

My own thoughts on using the Dominion engine are a little more along the lines of a different game than Dominion altogether, with a central mechanic using a variation of the deck building mechanism. I'm actually adapting an older idea I had to represent a players personal "tech tree" (set of abilities, like in Goa or Hansa Teutonica) and I think a deck building mechanism could work well. Each player's deck would contain cards relating to the various actions in the game, and whenever you take an action, you would add a card to your deck that associates with that action. The intensity/strength/potency of that action would be based on how many cards for that action are in your hand/drawn/revealed from the deck. Here's an example of my first thought for the mechanism:

Let's say there is an action in the game which is "collect wood" and another action which is "attack" - among other actions. Your starting deck would have maybe 2 cards for each action in it. When you choose to "collect wood," you reveal the top 2 cards from your deck and for each of those that has the "wood" symbol you collect 1 additional wood resource. Then you take a Collect Wood card and put it in your discard pile. In the future, your revealed cards are more likely to show wood icons because the concentration of them in your deck is higher. Similarly, if you choose the "attack" action, you would reveal 2 cards and get attack bonuses for each attack icon, then you would get a little better at attacking.

Now, consider that there are other cards you could add to your deck as well, perhaps an Axe. An Axe might have both a Wood symbol (because it helps you chop down trees and get wood), and an attack symbol (because it can be used as a weapon). Each time the Axe comes up, it will boost either the Attack or the Wood action, so it's better than either of the basic cards. The idea would be to get cards in your deck that boost the actions you want to take, and then of course take those actions. There would also be an action which allows you to reveal more cards at a time, thus making you 'better' at all actions - it would probably have some cost associated, allow you to take any 1 'basic' card, and increase the number of cards you reveal each time.

This 'tech tree' mechanism seems like it would work and be good for a variety of games, but thematically I think it would be nice in a civ building type of game where it represents the strength of your civilization in the various aspects of the game - population, warfare, agriculture, culture, technology... the kinds of things you find in games like Through the Ages. Michael had asked me to think about how a streamlined card game version of Twilight Imperium might work, and so I started considering this mechanism for something of that theme - a space opera civ-style card game. Here are my current thoughts on that:

Imagine there are 6 different actions in the game - probably the same or similar actions to those you'd see in Twilight Imperium. For each action there's a stack of cards in the center of the table, and each player begins the game with a deck of 2 cards for each of 5 the actions (we'll say that one action is called Warfare, and at the outset no one is good at that action so you don't start with any Warfare cards in your deck). There is also a deck of Planet cards which show the category or type of planet, and maybe some value as to how difficult it would be to settle, on the back of the card and details of the planet's benefits on the front. Finally, for each of the (3 or 4?) types of planets there is a deck of Technology cards - color coded so you know which associates with which.

Like Twilight Imperium, on your turn you select 1 action (role), and you take a card for that role from the center and add it to your discard pile (you have now become a little 'better' at that action for future turns). Then all players, starting with you, get a chance to take that action. Each player has a hand of cards, and when resolving the action you can play the appropriate cards from your hand to boost it in some way. Having chosen the role, perhaps the card you took from the center actually goes into your hand, effectively giving you an extra card of that type to use that turn. Possible actions are as follows...

Explore: Draw 1 card per Explore icon from the Planet deck LOOKING ONLY AT THE BACKS OF THE CARDS. Choose 1 of them and place it FACE DOWN in front of you. You have explored this planet, but have not settled it yet, and therefore get no benefit from it yet.

Settle: The back of the Planet card indicates how hard it is to Settle the planet. In order to succeed you must have the appropriate number of symbols. There could be different types of Planets -
- Uninhabited: In order to Settle you need a certain number of Settle icons.
- Civilized: In order to Settle you need a number of Settle icons, but you can use Armies to pay for some of them.
- Hostile: In order to Settle you must spend a certain number of Armies.

Research: You can research at a planet which you have settled. Draw 1 Technology card per Research symbol from the Technology deck matching a Planet you have Settled, and add 1 of those cards to your discard pile (putting the rest on the bottom of the deck). These cards would be like the Axe example above, they would be more efficient than the basic action cards, either giving more than 1 symbol of a type, or symbols of more than 1 type, improving your capabilities. The type of planet would indicate the sorts of abilities in the deck.

Warfare: Gain 1 Army per Warfare icon. These will be used to settle Hostile planets, and maybe could be used in a defensive capacity in some way.

Harvest: Produce 1 Resource per Harvest icon, perhaps based on the capabilities of planets that you have settled. These could be used to trade for VPs or perhaps could substitute for Armies or Settlement icons for Civilized (or Hostile?) planets. Perhaps some planets have slots on the back, and an alternate way to settle them is to fill those slots with the appropriate Resources.

Trade: Exchange 1 Resource per Trade icon for Victory Points

Turn by turn players would choose roles, allowing each player to participate in the role, and at the end (beginning?) of your turn you would refill your hand to 5 cards. Maybe you get a chance to discard cards you don't want, and maybe there's some way (via the technology deck) to remove cards in hand from the game so you don't need to draw them anymore. The game would end when either the Planet deck or one of the action piles, or possibly when the VP pool, is exhausted. Planets would be worth some VPs, as would points earned throughout the game by Trading*.

* Thought on trading that just popped into my head - perhaps when you trade stuff it goes into a common pool, and you can trade for other resources in the pool (pre-seeded with 1 or 2 of each) at a 1-for-1 rate. This would be a good way to get resources you don't have access to - which would matter if it took specific combinations of resources to do certain things. Maybe "victory points" is one of the "types of resources" - so you can always trade for those, or you can trade VPs for resources you might need to use (players would start with some VPs). The pool would be pre-seeded with a bunch of the VP resource, and maybe the game ends when that one runs out.

I think this could be a solid game using the deck building mechanism inspired by Dominion as a base for players to build a their strategy. How will they advance their game? Will they Settle a few select planets and then improve their technology, gaining points and abilities that way? Will they produce lots of Resources to trade? How will they settle planets - through normal means? Military force? Trade? I think this will offer players a lot of ways to go about improving their capabilities and earning points. I guess I'll have to make a prototype and find out if I'm right about that!


Isamoor said...

I didn't read all the details... but from what I did read, you seemed to be describing the basis of Race for the Galaxy. Of course, RftG is also a derivative of Puerto Rico, just like Twilight Imperium in many ways.

I mean, your suggested strategies are:

Produce Lots of Goods for Traiding
Settle Some Valuable planets

From your rambling, I see you went from "reveal the top 2" to "have a hand and play matching symbols". I would definitely think having a hand would add a lot of needed control.

Anyway, I think it could still work out. But I think it's straying pretty far from Dominion.

Hope it works out!

Seth Jaffee said...

Re: Race for the Galaxy
Yes, I noticed the similarity as soon as I added trade to the mix. However, I think my idea is pretty different from RftG in that there's a tech tree involved, so your incentives to pick one action over another might be way different tan in RftG.

I was considering removing trading and replacing it with some form of Politics... in some ways it would be similar, but at any rate instead of harvesting resources, you garner political support, and instead of shipping resources for VPs, maybe you spend the support to vote on agendas... if the agendas are basically just a way to get VPs then that's just a re-theming of trading (RftG) or shipping (PR). but if the agendas have some effect on the game, and/or if they're decided by a blind bid of Votes between players, then that could be really different. Of all the games I've seen which have a politics aspect, I like Warrior Knights' version the best, so I'd probably model any political aspect in my game after that (as described above). I've been shying away from politics thus far because I think forcing politics into a game kinda ruins the game for me (see TI3).

Re: reveal cards vs hand of cards
The original thoughts about using the deck building mechanism for a tech tree/ tech level indicator was to simply reveal some cards to see if your action get boosted. The idea was always that you could take whatever action you wanted, and when you did your deck would tune toward that action... and the boosts would be a sort of bonus that you might not be able to count on. The more you 'train' a certain action, the more (and more likely) boosts you'll get when you perform it. That would work well in a game where players take their own turns. In a game with a Role Selection mechanism, where everyone gets a chance to act on the role you chose on your turn, it seemed to make more sense to have a hand of cards, to inform your choice to follow the role or not. I suppose it could work either way, and in retrospect, like in Glory To Rome, if you don't want to do the action perhaps you should be able to draw a card as a consolation - thus making future actions better and/or giving you more options.

Re: Straying pretty far from Dominion
That's the point. Thunderstone is an example of "playing Dominion, only moreso" - an extension of the game Dominion. My thoughts are about making an entirely different game, where the main mechanism is inspired by the deck building mechanism from Dominion. I'm not trying to re-invent the wheel, Dominion does about as good a job as can be done. Thunderstone is neat, but as the designer said in his BGN article, it's more involved than Dominion - some players will like it for that reason, and some will hate it for that reason.

I'm certain my game, should it turn out to be good enough to pursue, will be compared to Dominion - and Race for the Galaxy, Puerto Rico, and maybe Twilight Imperium for that matter. But it's not intended to be something to play instead of those games. I'm hoping to fill a different niche - a streamlined civ building game with tech upgrades, like a light TI3, but more involved than RftG or San Juan. That's the target, I have o idea if this idea will actually hit it or not!

Jeff said...

I haven't played Dominion yet but the idea certainly sounds interesting. However, in your "TI light" example, it sounds like you're more or less representing with cards something that could just be tracked with a marker on a track. The idea itself seems better suited to a situation where you want to allow uncertain outcomes for actions, but where players build up their decks to improve their likelihood of succeeding at the actions they're most eager to succeed at.

This would work very well as a way to require specialization, since acquiring cards of all types just dilutes your probability of succeeding at any given action.

I could envision using an idea like this as a replacement for the achievement token system in Sands of Time.

-- When you take an action, draw a card (instead of receiving a token)

-- Having a Heritage card tells you what "degree" of card you get to draw; the better your Heritage card, the more symbols on your card

-- When you go to take an action that ordinarily costs tokens (complete an Advance, record a Chronicle, or pay tokens in a battle), instead flip X cards and see whether you have enough symbols. Or maybe instead, flip X cards and then decide what you want to do based on the symbols you flipped.

But there has to be some way to keep track of what X ought to be.

I guess the advantage of such as system would be that it could fit well with the theme -- since the cards would represent your empire's reputation, the fact that you get to keep them makes more thematic sense than tokens, which are spent. Revealing cards randomly reflects the imperfection of communication in those days; completing an action requires not just that you've done something, but that people have heard about it, which is somewhat outside your control (although maybe you can influence it). And, you are encouraged to specialize to have greater success at pulling the symbols you need for the actions you want to emphasize. (eg, if you're following a "war" strategy, you want to get a lot of political cards to increase your chances of pulling political cards from the deck).

The card-flipping might inject too much randomness into a long, "serious" game like Sands, but it's at least fun to think about it.

Seth Jaffee said...

The original idea for this 'tech tree' format was inspired by a video game (which I've never played) called Dungeon Siege. As I understand it, in that game your character doesn't have a specific "class" like in most RPGs. Instead, you can do whatever you want - swing a sword, cast a spell, heal yourself, pick a lock... the more you swing a sword, the better you get at swinging a sword, which maybe encourages you to swing swords more often, and before you know it your character has the attributes of a fighter. Similarly, as you cast spells, the better you get at casting spells (the more powerful your spells become). Ditto shooting arrows, healing, etc.

I've been itching to use this mechanism (wherein the more you do an action the better you get at doing that action) in a game for quite some time now. This new idea is my first real attempt to do that.

Seth Jaffee said...

"The card-flipping might inject too much randomness into a long, "serious" game like Sands"

Jeff - I actually think the way a deck of cards will behave is actually a really good fit for a longer civ game like Sands - it will give time for the deck to grow and cycle through several times at each 'level'... early game there would be a generally even mix of cards so your starting empire would be generally the same as everyone else's. For variable setup, each player could start with a different deck composition (+1 card of a certain type, -1 of a different type).

As the game goes on and actions are chosen by players, their decks start to grow and the composition changes, favoring the actions they've taken so far. Later in the game, presumably, the value of X (the number of cards revealed) will increase - either by rule ("Reveal 2/3/4/5 cards in Era I/II/III/IV") or because people will upgrade it on their own via some game action. Later in the game, a focused deck will really be good at doing whatever action it is focused on, and a player could probably specialize in 2 things heavily and late in the game when their deck has a high concentration of those actions and they're revealing several cards they can reasonably expect a potent action. A more well rounded civ wouldn't have as powerful an action as the focused civ, but would be able to do a variety of actions with reasonable strength.

If you'd like to try such a system for Sands, then I'd be interested to hear how it goes. In my game I'm going to have planets that players will "Settle" or colonize, and once they do, the planet may give them an additional, permanent +1 virtual symbol on a particular action. A similar thing could be used in Sands - when you construct a monument or develop a particular thing, it could have a permanent effect which amounts to +1 symbol toward a particular action, thus helping that action EVERY time you do it, rather than tuning your deck toward that action.

Anonymous said...

Your blog keeps getting better and better! Your older articles are not as good as newer ones you have a lot more creativity and originality now keep it up!

Seth Jaffee said...

Um... thanks, I guess!